CN105760399A - Data retrieval method and device - Google Patents

Abstract

The embodiment of the invention discloses a data retrieval method and equipment, which relate to the field of communication technology and are used to improve the accuracy of retrieval results and improve retrieval efficiency. The specific solution is: the call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text; obtains the first word segmentation set identified by text analysis of the text information, and the first word segmentation set contains at least A text word segmentation; from the keyword index list of the retrieval data stored in the knowledge database, search for the retrieval keyword matching the text word segmentation in the first word segmentation set, the keyword index list contains at least one keyword index item, keyword The index item includes the search keyword and the identifier of the search data corresponding to the search keyword; from the knowledge database, the search data indicated by the identifier of the retrieved data corresponding to the found search keyword is retrieved. The invention is used in data retrieval process.

Description

A data retrieval method and device

technical field

The invention relates to the field of communication technology, in particular to a data retrieval method and device.

Background technique

With the rapid development of informatization, the amount of information reserves in various industries or fields is also increasing. For example, in order to be able to provide users with more comprehensive, fast and accurate services such as business consultation and tariff inquiry, an operator that provides communication services to users has built a knowledge database that can be retrieved by the customer service personnel of the telecom operator.

Among them, after the customer service personnel connect to the voice communication with the user through the call center server, they can identify the search keywords from the user voice information according to their personal business experience and skills; then input the search keywords into the knowledge database, and the knowledge database Match the information that matches the search keyword as the search data, and display the search data to the customer service personnel; the customer service personnel also need to extract information that is highly correlated with the user's voice information from the search data displayed in the knowledge database based on their personal business experience and skills. Retrieve data, and respond to the user according to the extracted retrieval data, or feed back the extracted retrieval data to the user.

However, in the above data retrieval process, when the customer service personnel retrieve from the knowledge database, if they want to quickly retrieve the retrieved data with a high degree of matching and correlation with the user's voice information, it is necessary for the customer service personnel to have rich personal business knowledge. Experience and high personal business skills, the accuracy of the search results depends too much on the personal business experience and skills of the customer service personnel; and, the customer service personnel will inevitably make some retrieval errors due to negligence or carelessness during the retrieval process, then It will affect the accuracy and retrieval efficiency of retrieval, resulting in low accuracy and retrieval efficiency of retrieval results.

Contents of the invention

Embodiments of the present invention provide a data retrieval method and device to improve the accuracy of retrieval results and improve retrieval efficiency.

In order to achieve the above object, embodiments of the present invention adopt the following technical solutions:

The first aspect of the embodiments of the present invention provides a data retrieval method, including:

The call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text;

The call center server acquires a first word segmentation set identified by performing text analysis on the text information, and the first word segmentation set includes at least one text segmentation word;

The call center server searches for a retrieval keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, and the keyword index list contains at least one keyword An index item, the keyword index item includes a search keyword and an identifier of the search data corresponding to the search keyword, and at least one search data and the identifier of the search data are stored in the knowledge database;

The call center server retrieves, from the knowledge database, the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword.

With reference to the first aspect, in a first possible implementation manner, in the keyword index items in the keyword index list of the retrieved data stored by the call center server in the knowledge database, search Before the search keyword matched by the text word segmentation, the method also includes:

The call center server reads the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database;

The call center server obtains a second word segmentation set identified by performing text analysis on the first retrieval data, and the second word segmentation set includes at least one text segmentation word;

The call center server generates the retrieval keywords of the first retrieval data according to the text segmentation in the second segmentation set;

The call center server generates one keyword index item according to the retrieval keyword of the first retrieval data and the identifier of the first retrieval data, and generates the keyword index containing at least one keyword index item list.

With reference to the first aspect, in a second possible implementation manner, in the keyword index items in the keyword index list of the retrieved data stored by the call center server in the knowledge database, search for items related to the first Before the retrieval keyword matched by the text word segmentation in the word segmentation set, the method also includes:

The keyword index list sent by the text analysis server is received.

With reference to any one of the first aspect, the first possible implementation manner, and the second possible implementation manner, in a third possible implementation manner, the call center server obtains and performs text analysis on the text information The recognized first participle set, including:

The call center server performs text analysis on the text information, and recognizes a third participle set, which contains at least one text participle;

The call center server filters the meaningless text participles included in the third participle set to obtain the first participle set, and the meaningless text participle includes: exclamation words and modal particles.

With reference to any one of the first aspect, the first possible implementation manner, and the second possible implementation manner, in a fourth possible implementation manner, the call center server obtains and performs text analysis on the text information The recognized first participle set, including:

The call center server performs text analysis on the text information, and recognizes a third participle set, which contains at least one text participle;

The call center server filters the first text segmentation included in the third word segmentation set to obtain the first text segmentation set, and the first text segmentation is the combination of the third word segmentation set and the keyword index list A text segment that does not match any of the search keywords.

With reference to the first aspect or the second possible implementation manner, in a fifth possible implementation manner, the speech information is composed of at least one speech segment whose dimension is a sentence, and the text information is composed of at least one text segment;

The call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text, including:

The call center server receives the voice segment sent by the user terminal;

The call center server sends the voice segment to the voice analysis server after receiving one of the voice segments, so that the voice analysis server converts the voice segment into the text segment;

The call center server receives the text segment sent by the speech analysis server.

With reference to the first aspect or the second possible implementation manner, in a sixth possible implementation manner, the call center server obtains the first word segmentation set identified by performing text analysis on the text information, including:

The call center server sends the text information to the text analysis server, so that the text analysis server performs text analysis on the text information and identifies the first word segmentation set;

The call center server receives the first word segmentation set sent by the text analysis server.

A second aspect of the embodiments of the present invention provides a data retrieval method, including:

The text analysis server receives the text information sent by the call center server, and the text information is converted by the call center server according to the voice information;

The text analysis server performs text analysis on the text information, and identifies the first word segment set, and the first word segment set contains at least one text segment word;

The text analysis server sends the first word segmentation set to the call center server, so that the call center server retrieves data from the knowledge database according to the text segmentation words in the first word segmentation set.

With reference to the second aspect, in a first possible implementation manner, at least one search data and an identifier of the search data are stored in the knowledge database;

The method also includes:

The text analysis server reads the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database;

The text analysis server performs text analysis on the first retrieval data, and identifies a second word segmentation set, and the second word segmentation set contains at least one text segmentation word;

The text analysis server generates the retrieval keywords of the first retrieval data according to the text segmentation in the second segmentation set;

The text analysis server generates a keyword index item according to the search keyword of the first search data and the identifier of the first search data, and generates the keyword index containing at least one keyword index item list;

The text analysis server sends the keyword index list to the call center server.

A third aspect of the embodiments of the present invention provides a call center server, including:

a receiving unit, configured to receive voice information sent by the user terminal;

a text conversion unit, configured to obtain text information obtained by performing text conversion on the voice information received by the receiving unit;

a text analysis unit, configured to obtain a first word segmentation set identified by performing text analysis on the text information obtained by the text conversion unit, and the first word segmentation set includes at least one text segment;

A keyword search unit, configured to search for a search keyword that matches the text word segmentation in the first word segmentation set obtained by the text analysis unit from the keyword index list of the retrieval data stored in the knowledge database, the keyword The word index list contains at least one keyword index item, and the keyword index item includes a search keyword and an identification of the search data corresponding to the search keyword, and at least one search data and the search data are stored in the knowledge database. identification of the retrieved data;

The retrieval unit is configured to retrieve, from the knowledge database, the retrieval data indicated by the identifier of the retrieved data corresponding to the retrieval keyword found by the keyword searching unit.

With reference to the third aspect, in a first possible implementation manner, the call center server may further include:

A reading unit, configured to search for, among the keyword index items in the keyword index list of the retrieval data stored in the knowledge database by the keyword search unit, the words that match the text word segmentation in the first word segmentation set Before retrieving keywords, read the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database;

The text analysis unit is further configured to obtain a second word segmentation set identified by performing text analysis on the first retrieval data read by the reading unit, and the second word segmentation set includes at least one text segmentation word;

A generation unit, configured to generate a retrieval keyword of the first retrieval data according to the text word segmentation in the second word segmentation set obtained by the text analysis unit; according to the retrieval keyword of the first retrieval data and the The identification of the first retrieved data generates a key index item.

With reference to the third aspect, in a second possible implementation manner, the receiving unit is further configured to use the keyword index in the keyword index list of the retrieved data stored in the knowledge database by the keyword searching unit In the item, before searching for the retrieval keyword matching the text word segmentation in the first word segmentation set, the keyword index list sent by the text analysis server is received.

In combination with any one of the third aspect, the first possible implementation manner, and the second possible implementation manner, in a third possible implementation manner, the text analysis unit includes:

A text analysis module, configured to perform text analysis on the text information, and identify a third word segment set, the third word segment set includes at least one text segment word;

The first filtering module is used to filter the nonsense text participles contained in the third participle set identified by the text analysis module to obtain the first participle set, and the nonsense text participles include: exclamation words and tone word.

In combination with any one of the third aspect, the first possible implementation manner, and the second possible implementation manner, in a fourth possible implementation manner, the text analysis unit includes:

A text analysis module, configured to perform text analysis on the text information, and identify a third word segment set, the third word segment set includes at least one text segment word;

The second filtering module is configured to filter the first text segmentation contained in the third segmentation set identified by the text analysis module to obtain the first segmentation set, and the first text segmentation is the third segmentation Text word segmentation in the collection that does not match any search keyword in the keyword index list.

With reference to the third aspect or the second possible implementation manner, in a fifth possible implementation manner, the speech information is composed of at least one speech segment whose dimension is a sentence, and the text information is composed of at least one text segment;

The receiving unit is specifically configured to receive the voice segment sent by the user terminal;

The text conversion unit is specifically configured to, after the receiving unit receives a voice segment, send the voice segment to a voice analysis server, so that the voice analysis server converts the voice segment into the text Fragment: receiving the text fragment sent by the speech analysis server.

With reference to the third aspect or the second possible implementation manner, in a sixth possible implementation manner, the text analysis unit is specifically configured to send the text information to the text analysis server, so that the text analysis The server performs text analysis on the text information to identify the first word segmentation set; and receives the first word segmentation set sent by the text analysis server.

A fourth aspect of the embodiments of the present invention provides a text analysis server, including:

a receiving unit, configured to receive text information sent by the call center server, the text information being converted by the call center server according to the voice information;

A text analysis unit, configured to perform text analysis on the text information received by the receiving unit, and identify the first word segment set, the first word segment set includes at least one text segment word;

A sending unit, configured to send the first word segmentation set analyzed by the text analysis unit to the call center server, so that the call center server retrieves from the knowledge database according to the text word segmentation in the first word segmentation set data.

With reference to the fourth aspect, in a first possible implementation manner, at least one retrieval data and an identifier of the retrieval data are stored in the knowledge database;

The text analysis server also includes:

A reading unit, configured to read first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database;

The text analysis unit is further configured to perform text analysis on the first retrieval data read by the reading unit, and identify a second word segment set, the second word segment set includes at least one text segment word;

A generating unit, configured to generate a retrieval keyword of the first retrieval data according to the text segmentation in the second segmentation set analyzed by the text analysis unit; according to the retrieval keyword of the first retrieval data and the The identification of the first retrieval data generates one of the keyword index items, and generates the keyword index list including at least one of the keyword index items;

The sending unit is further configured to send the keyword index list generated by the generating unit to the call center server.

In the data retrieval method and device provided by the embodiments of the present invention, the call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text; A set of word segments, the first set of word segments contains at least one text segment; the call center server searches for the search keywords that match the text segments in the first set of word segments from the keyword index list of the retrieval data stored in the knowledge database, and the key The word index list includes at least one keyword index item, the keyword index item includes a search keyword and an identifier of the search data corresponding to the search keyword, and at least one search data and the identifier of the search data are stored in the knowledge database; The call center server retrieves, from the knowledge database, the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword.

Compared with the prior art when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid search errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Description of drawings

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the following will briefly introduce the drawings that need to be used in the description of the embodiments or the prior art. Obviously, the accompanying drawings in the following description are only These are some embodiments of the present invention. For those skilled in the art, other drawings can also be obtained according to these drawings without any creative effort.

Fig. 1 is a flow chart of a data retrieval method provided by Embodiment 1 of the present invention;

Fig. 2 is a flow chart of a data retrieval method provided by Embodiment 2 of the present invention;

FIG. 3 is a flowchart of a data retrieval method provided by Embodiment 3 of the present invention;

FIG. 4 is a flowchart of a data retrieval method provided by Embodiment 4 of the present invention;

FIG. 5 is a schematic diagram of the structural composition of a call center server provided in Embodiment 5 of the present invention;

FIG. 6 is a schematic diagram of the structural composition of another call center server provided by Embodiment 5 of the present invention;

FIG. 7 is a schematic diagram of the structure and composition of another call center server provided by Embodiment 5 of the present invention;

FIG. 8 is a schematic diagram of the structure and composition of another call center server provided by Embodiment 5 of the present invention;

FIG. 9 is a schematic diagram of the structural composition of a text analysis server provided in Embodiment 6 of the present invention;

FIG. 10 is a schematic diagram of the structural composition of another text analysis server provided by Embodiment 6 of the present invention;

FIG. 11 is a schematic diagram of the structural composition of a call center server provided by Embodiment 7 of the present invention;

FIG. 12 is a schematic diagram of the structural composition of another call center server provided by Embodiment 7 of the present invention;

FIG. 13 is a schematic diagram of the structural composition of another call center server provided by Embodiment 7 of the present invention;

FIG. 14 is a schematic diagram of the structural composition of a text analysis server provided in Embodiment 8 of the present invention;

FIG. 15 is a schematic diagram of the structure and composition of a data retrieval system provided by Embodiment 8 of the present invention.

detailed description

The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some, not all, embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by persons of ordinary skill in the art without making creative efforts belong to the protection scope of the present invention.

The various technologies described herein can be used in various wireless communication systems, such as current 2G, 3G communication systems and next-generation communication systems, such as Global System for Mobile Communications (GSM, Global System for Mobile communications), Code Division Multiple Access (CDMA, CodeDivisionMultipleAccess) system, Time Division Multiple Access (TDMA, TimeDivisionMultipleAccess) system, Wideband Code Division Multiple Access (WCDMA, WidebandCodeDivisionMultipleAccessWireless), Frequency Division Multiple Access (FDMA, FrequencyDivisionMultipleAddressing) system, Orthogonal Frequency Division Multiple Access (OFDMA, OrthogonalFrequency-DivisionMultipleAccess) system, single carrier FDMA (SC-FDMA) system, General Packet Radio Service (GPRS, General Packet Radio Service) system, Long Term Evolution (LTE, Long Term Evolution) system, and other such communication systems.

Various aspects are described herein in connection with terminals and/or base stations and/or base station controllers.

The user terminal may be a wireless terminal or a wired terminal. The wireless terminal may be a device that provides voice and/or data connectivity to the user, a handheld device with a wireless connection function, or other processing devices connected to a wireless modem. The wireless terminal can communicate with one or more core networks via a radio access network (for example, RAN, RadioAccessNetwork), and the wireless terminal can be a mobile terminal, such as a mobile phone (or called a "cellular" phone) and a computer with a mobile terminal , for example, may be portable, pocket, handheld, computer built-in, or vehicle-mounted mobile devices that exchange voice and/or data with the radio access network. For example, Personal Communication Service (PCS, Personal Communication Service) phone, cordless phone, Session Initiation Protocol (SIP) phone, Wireless Local Loop (WLL, Wireless Local Loop) station, Personal Digital Assistant (PDA, Personal Digital Assistant) and other equipment. Wireless terminal can also be called system, subscriber unit (SubscriberUnit), subscriber station (SubscriberStation), mobile station (MobileStation), mobile station (Mobile), remote station (RemoteStation), access point (AccessPoint), remote terminal (RemoteTerminal) , an access terminal (AccessTerminal), a user terminal (UserTerminal), a user agent (UserAgent), a user equipment (UserDevice), or a user equipment (UserEquipment).

Additionally, the terms "system" and "network" are often used herein interchangeably. The term "and/or" in this article is just an association relationship describing associated objects, which means that there can be three relationships, for example, A and/or B can mean: A exists alone, A and B exist simultaneously, and there exists alone B these three situations. In addition, the character "/" in this article generally indicates that the contextual objects are an "or" relationship.

The method of the embodiment of the present invention is applied in the call center system, and the call center system makes full use of modern communication and computer technology, such as interactive voice response (English: InteractiveVoiceResponse, abbreviated: IVR), automatic call distribution (English: AutomaticCallDistributor, abbreviated: ACD) and other technologies can automatically and flexibly handle a large number of different telephone inbound and outbound business and operational operating systems.

Among them, the existing call center system is mainly composed of a call center server and a knowledge database. Take the call center system of a communication operator (operator that provides communication services for users) as an example to process user inquiry calls. After the customer service staff connects to the voice communication with the user through the call center server in the call center system, they can The business experience and skills of the user identify the search keywords from the user's voice information; then enter the search keywords into the knowledge database, and the knowledge database matches the information that matches the search keywords as the search data, and displays the search data to the customer service personnel; Personnel also need to extract retrieval data that is highly correlated with user voice information from the retrieval data displayed in the knowledge database based on personal business experience and skills, and respond to users based on the extracted retrieval data, or use the extracted retrieval data to The data is fed back to the user.

Different from the prior art, when the call center system in the embodiment of the present invention handles user inquiry calls, the call center server can realize text conversion of voice information and text analysis of text information instead of customer service personnel. Information identification and analysis in the process of processing user queries can improve the accuracy of retrieval results and improve retrieval efficiency. The data retrieval method provided by the present invention and the equipment for implementing the data retrieval method are now described with the following specific examples:

Embodiment one

An embodiment of the present invention provides a data retrieval method, as shown in Figure 1, including:

S101. The call center server receives voice information sent by a user terminal, and acquires text information obtained by performing text conversion on the voice information.

Wherein, the user may initiate a voice communication request to the call center server through the user terminal, so as to send voice information to the call center server after establishing a voice communication link between the user terminal and the call center server. Wherein, the voice information sent by the user terminal is composed of at least one voice segment whose dimension is a sentence.

It should be noted that the call center server in the embodiment of the present invention may be a call center management server that receives calls from users from operators that can provide communication services to users.

S102. The call center server acquires a first word segmentation set identified through text analysis of the text information, where the first word segmentation set includes at least one text segmentation word.

Wherein, the method for the call center server to obtain the first word segmentation set identified by text analysis of the text information may specifically include: the call center server performs sentence segmentation and word segmentation on the text information to obtain at least one text segment.

Exemplarily, the call center server can perform text analysis on the text information, identify the third word segmentation set (the third word segmentation set contains at least one text segmentation word), and then filter the meaningless text segmentation words contained in the third word segmentation set to obtain the first A set of participles, meaningless text participles include: exclamation words and modal particles.

Exemplarily, the call center server performs text analysis on the text information, identifies the third word segmentation set (the third word segmentation set contains at least one text segmentation word), and then filters the first text segmentation word contained in the third word segmentation set to obtain the first The word segmentation set, the first text word segmentation is the text word segmentation in the third word segmentation set that does not match any search keyword in the keyword index list.

Further, in an application scenario of the embodiment of the present invention, the call center server can send text information to the text analysis server, and the text analysis server performs text analysis on the text information to identify the first word segmentation set; then the text analysis server receives The set of first participles sent.

S103. The call center server searches for a search keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the search data stored in the knowledge database.

Wherein, the keyword index list includes at least one keyword index item, and the keyword index item includes a retrieval keyword and an identification of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identification of the retrieval data are stored in the knowledge database .

In an application scenario of the embodiment of the present invention, the keyword index item may be that the call center server performs text analysis on each retrieval data stored in the knowledge database, and the retrieval keyword of each retrieval data obtained corresponds to the retrieval keyword The identity of the retrieved data constitutes.

Wherein, the call center server performs text analysis on each retrieval data stored in the knowledge database, and the specific method for obtaining the retrieval keywords of each retrieval data may include: the call center server reads the first retrieval data stored in the knowledge database , the first search data is any search data stored in the knowledge database; obtain a second word segmentation set identified by text analysis on the first search data, and the second word segmentation set contains at least one text Word segmentation: generating a search keyword for the first search data according to the text word segmentation in the second word segmentation set.

Furthermore, in another application scenario of the embodiment of the present invention, the call center server may receive the keyword index list sent by the text analysis server. Wherein, the method for generating the keyword index list by the text analysis server is similar to the method for generating the keyword index list by the call center server, which will not be repeated here in this embodiment.

S104. The call center server searches the knowledge database for the search data indicated by the identifier of the search data corresponding to the found search keyword.

Wherein, after the call center server finds the retrieval keyword matching the text word segmentation in the first word segmentation set, it can retrieve the retrieval data indicated by the identifier of the retrieved data corresponding to the retrieval keyword found from the knowledge database. data.

In the data retrieval method provided by the embodiment of the present invention, the call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text; the call center server obtains the first participle recognized by text analysis of the text information Set, the first word segmentation set contains at least one text segmentation word; the call center server searches the search keyword matching the text segmentation word in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, keyword index The list contains at least one keyword index item, and the keyword index item includes a retrieval keyword and an identification of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identification of the retrieval data are stored in the knowledge database; the call center The server retrieves the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword from the knowledge database.

Compared with the prior art when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid search errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Embodiment two

An embodiment of the present invention provides a data retrieval method, as shown in Figure 2, including:

S201. The text analysis server receives the text information sent by the call center server, where the text information is converted by the call center server according to the voice information.

Wherein, the text information is composed of at least one text segment, and the speech information is composed of at least one speech segment with a sentence as a dimension.

The call center server can send a voice segment to the voice analysis server after receiving a voice segment sent by the user terminal during voice communication with the user terminal, so that the voice analysis server converts the voice segment into a text segment; call After receiving the text segment sent by the voice analysis server, the central server may send the text segment to the text analysis server.

S202. The text analysis server performs text analysis on the text information, and identifies a first word segment set, where the first word segment set includes at least one text segment word.

Wherein, the method for the text analysis server to perform text analysis on the text information to identify the first word segmentation set may specifically include: the text analysis server performs sentence segmentation and word segmentation on the text information to identify at least one text segmentation.

Exemplarily, the text analysis server can perform text analysis on the text information, identify the third word segmentation set (the third word segmentation set contains at least one text segmentation word), and then filter the meaningless text segmentation words contained in the third word segmentation set to obtain the first A set of participles, meaningless text participles include: exclamation words and modal particles.

Exemplarily, the text analysis server may perform text analysis on the text information, identify a third word segment set (the third word segment set contains at least one text segment word), and then filter the first text segment words contained in the third word segment set to obtain the first text segment A word segmentation set, the first text segmentation is a text segmentation that does not match any search keyword in the keyword index list in the third segmentation set.

S203. The text analysis server sends the first word segmentation set to the call center server, so that the call center server retrieves data from the knowledge database according to the text segmentation words in the first word segmentation set.

Further, the text analysis server can perform text analysis on each retrieval data stored in the knowledge database to obtain the retrieval keyword of each retrieval data, and then generate a text file containing the retrieval keyword of the first retrieval data and the identifier of the first retrieval data. Keyword index items, and generate a keyword index list containing at least one keyword index item, and then send the keyword index list to the call center server, so that the call center server can retrieve the keyword index list of data stored in the knowledge database , search for a search keyword matching the text word segmentation in the first word segmentation set, and retrieve the search data indicated by the identifier of the retrieved data corresponding to the found search keyword from the knowledge database.

It can be understood that the method of this embodiment may also include: the text analysis server reads the first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database; Text analysis is performed on the first retrieval data, and a second word segmentation set is identified, which contains at least one text segmentation word; the text analysis server generates a retrieval keyword of the first retrieval data according to the text segmentation words in the second word segmentation set; the text The analysis server generates a keyword index item according to the retrieval keyword of the first retrieval data and the identification of the first retrieval data; the text server generates a keyword index list containing at least one keyword index item; the text analysis server sends the key to the call center server List of word indexes.

In the data retrieval method provided by the embodiment of the present invention, the text analysis server receives the text information sent by the call center server, and the text information is converted by the call center server according to the voice information; the text analysis server performs the text information on the text information Text analysis, identifying the first word segmentation set, the first word segmentation set contains at least one text segmentation word; the text analysis server sends the first word segmentation set to the call center server, so that the call center The server retrieves data from the knowledge database according to the text word segmentation in the first word segmentation set.

Compared with the prior art, when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the text analysis server can query the call center server The sent text information is subjected to text analysis to obtain the analysis result (the first word segmentation set), and the first word segmentation set is sent to the call center server, and the call center server searches the keyword index list of the retrieval data stored in the knowledge database, and The search keywords matched by the text word segmentation in the first word segmentation set do not require the customer service personnel to identify the search keywords from the user voice information based on their personal business experience and skills, and the call center server automatically performs the recognition and comparison of the search keywords Because the customer service personnel identify the retrieval keywords based on their personal business experience and skills, it can improve the recognition efficiency of the retrieval keywords, avoid retrieval errors due to negligence or carelessness of the customer service personnel, and improve the accuracy of identifying retrieval keywords. After the search keywords are obtained efficiently and accurately, the search results can be obtained quickly and accurately.

Embodiment Three

An embodiment of the present invention provides a data retrieval method, which is applied when the call center server is integrated with a text conversion module capable of performing text conversion on speech information and capable of performing text conversion on the text information obtained by text conversion of the speech information by the text conversion module. The analyzed text analysis module, that is, the call center server can perform text conversion on speech information, and then perform text analysis on the text information obtained through text conversion in a specific scenario.

In this application scenario, as shown in Figure 3, the data retrieval method may include:

S301. The call center server reads first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database.

S302. The call center server acquires a second word segmentation set identified by performing text analysis on the first retrieval data, and the second word segmentation set includes at least one text segmentation word.

Wherein, the call center server may perform text analysis on any retrieval data stored in the knowledge database, and identify a second word segmentation set, and the second word segmentation set includes at least one text segmentation word. The method for the call center server to perform text analysis on the retrieval data stored in the knowledge database is similar to the method for the call center server to perform text analysis on the text information, and the method for the call center server to perform text analysis on the retrieval data stored in the knowledge database can refer to the present invention The specific method for the text analysis performed by the call center server on the text information in the embodiment will not be repeated here in the embodiment of the present invention.

S303. The call center server generates a search keyword of the first search data according to the text word segmentation in the second word segmentation set.

S304. The call center server generates a keyword index item according to the search keyword of the first search data and the identifier of the first search data.

S305. The call center server generates a keyword index list including at least one keyword index item.

Exemplarily, the embodiment of the present invention takes Table 1 as an example to provide an example of a keyword index list:

Table 1:

S306. The call center server receives the voice information sent by the user terminal, and acquires text information obtained by performing text conversion on the voice information.

Wherein, the call center server may perform text conversion on the voice information to obtain text information. The method that the call center server can perform text conversion on voice information to obtain text information can refer to the specific method for realizing voice conversion to text in the prior art, which will not be repeated here in the embodiment of the present invention.

Exemplarily, an application program capable of converting speech to text may be installed in the call center server, so as to realize the conversion from speech information to text information.

S307. The call center server acquires a first word segmentation set identified by performing text analysis on the text information, where the first word segmentation set includes at least one text segmentation word.

Wherein, the method for the call center server to obtain the first word segmentation set identified by text analysis of the text information may specifically include: the call center server performs sentence segmentation and word segmentation on the text information to obtain at least one text segment.

Exemplarily, after the call center server obtains the text information obtained by converting the voice information into text: "Please check the tariff of package A for me", then the text information: "Please check the tariff of package A for me." Tariff situation” is segmented and word-segmented, and the first participle set consisting of text segmentation: “cha”, “package A” and “tariff situation” can be obtained.

Furthermore, after the call center server performs text analysis on the text information to obtain the text segmentation, it can also filter out meaningless text segmentation in the analyzed text segmentation, so as to improve data retrieval efficiency. Wherein, the nonsense text participle in the embodiment of the present invention may include: exclamation words and modal particles; the nonsense text participle may also include greetings and polite expressions.

Exemplarily, suppose that the text message obtained by the call center server is "Hi, please help me check the tariff of package A, thank you", then the text message can be: "Hello, please help me check the package A The tariff situation of A, thank you" can be segmented and word-segmented, and the text segmentation can be obtained: "Hello" (greeting), "please" (polite language), "check", "package A", "rate situation ", "Thank you" (polite expressions) and "啦" (exclamation words/modal particles) form the third participle set; the call center server can filter the greetings in the third participle set: "Hello", polite words: " Please" and "thank you" and the exclamation/modal particle: "la", get the first participle set consisting of the text participle: "cha", "package A" and "tariff".

Further, the call center server can also search the keyword index list for a search keyword that matches each text word in the third word segmentation set after obtaining the third word segmentation set; if any of the keyword index lists The search keywords do not match a text segment in the third segment set, and the call center server can filter the text segment.

The method of this embodiment may also include: the call center server performs text analysis on the text information, identifies a third word segmentation set (at least one text segmentation word is included in the third word segmentation set), and then filters the first text contained in the third word segmentation set Word segmentation, to obtain the first word segmentation set, the first text segmentation is the text segmentation that does not match any search keyword in the keyword index list in the third word segmentation set.

Based on the above example, assuming that any search keyword in the keyword index list does not match the text word segmentation: "check", the call center server can filter the text word segmentation: "check", and obtain the text word segmentation: "package A" and The first participle set composed of "tariff situation".

It should be noted that, in the embodiment of the present invention, S301-S305 may be executed first, and then S306-S307 may be executed; or S306-S307 may be executed first, and then S301-S305 may be executed. The embodiment of the present invention does not limit the sequence of execution of S301-S305 and S306-S307.

S308. The call center server searches for a search keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the search data stored in the knowledge database.

S309. The call center server searches the knowledge database for the search data indicated by the identifier of the search data corresponding to the found search keyword.

Wherein, at least one retrieval data and an identification of the retrieval data are stored in the knowledge database.

Exemplarily, the embodiment of the present invention takes Table 2 as an example, the knowledge database can store a plurality of retrieval data and the identification of each retrieval data in the form of a list:

Table 2:

In the data retrieval method provided by the embodiment of the present invention, the call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text; the call center server obtains the first participle recognized by text analysis of the text information Set, the first word segmentation set contains at least one text segmentation word; the call center server searches the search keyword matching the text segmentation word in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, keyword index The list contains at least one keyword index item, and the keyword index item includes a retrieval keyword and an identification of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identification of the retrieval data are stored in the knowledge database; the call center The server retrieves the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword from the knowledge database.

Compared with the prior art when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid search errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Embodiment Four

The embodiment of the present invention provides a data retrieval method, which is applied in a specific scenario where the call center server does not have the functions of performing text conversion on voice information and text analysis on text information.

In this application scenario, as shown in Figure 4, the data retrieval method may include:

S401. The text analysis server reads first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database.

Exemplarily, the knowledge database can store a plurality of retrieval data and the identification of each retrieval data in the form of a list. For the specific form and content of the knowledge database, refer to the list form example of the knowledge database shown in Table 2 of the embodiment of the present invention.

S402. The text analysis server performs text analysis on the first retrieval data, and identifies a second word segment set, where the second word segment set includes at least one text segment.

Wherein, the first retrieval data is any retrieval data stored in the knowledge database. The text analysis server can perform text analysis on any retrieval data stored in the knowledge database, and identify a second word segmentation set, which contains at least one text segmentation word. The method for the text analysis server to perform text analysis on the retrieval data stored in the knowledge database is similar to the method for text analysis performed by the call center server on text information, and the method for the text analysis server to perform text analysis on the retrieval data stored in the knowledge database can refer to the present invention The specific method for the text analysis performed by the call center server on the text information in the embodiment will not be repeated here in the embodiment of the present invention.

S403. The text analysis server generates a search keyword of the first search data according to the text word segmentation in the second word segmentation set.

S404. The text analysis server generates a keyword index item according to the search keyword of the first search data and the identifier of the first search data.

S405. The text server generates a keyword index list including at least one keyword index item, and sends the keyword index list composed of at least one keyword index item to the call center server.

For the specific form and content of the keyword index list, refer to the keyword index list shown in Table 1 of the embodiment of the present invention. In addition, for the method for the text server to generate the keyword index list, reference may be made to the specific description of the method for the call center server to generate the keyword index list in the embodiment of the present invention, and details will not be repeated here in the embodiment of the present invention.

S406. The call center server receives the voice information sent by the user terminal, and acquires text information obtained by performing text conversion on the voice information.

Wherein, the voice information is composed of at least one voice segment with a sentence as a dimension.

As shown in Figure 4, S406 may specifically include: S406a-S406d:

S406a. The call center server receives the voice segment sent by the user terminal.

S406b. After receiving a voice segment, the call center server sends the voice segment to the voice analysis server.

S406c. The speech analysis server converts the speech segment into a text segment.

Wherein, the method for the voice analysis server to convert the voice segment into the text segment is similar to the method for the call center server to convert the voice segment into the text segment, which will not be repeated here in the embodiment of the present invention.

S406d. The call center server receives the text segment sent by the speech analysis server.

S407. The call center server sends the text information to the text analysis server.

Wherein, the text information is composed of at least one text fragment, and the call center server may send the text fragment to the text analysis server after receiving a text fragment sent by the speech analysis server; of course, the call center server may also receive the speech analysis After sending all the text fragments constituting the text information, the server sends the text information to the text analysis server.

S408. The text analysis server performs text analysis on the text information, and identifies the first word segmentation set.

It should be noted that, the text analysis server performs text analysis on the text information, and the method for identifying the first word segmentation set can refer to the specific method for the call center server to perform text analysis on the text information, and identify the first word segmentation set, the embodiment of the present invention is here No longer.

S409. The text analysis server sends the first word segmentation set to the call center server.

It should be noted that, in the embodiment of the present invention, S401-S405 may be executed first, and then S406-S409 may be executed; or S406-S409 may be executed first, and then S401-S405 may be executed. In this embodiment of the present invention, there is no limitation on the execution sequence of S401-S405 and S406-S409.

S410. The call center server searches for a search keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the search data stored in the knowledge database.

S411. The call center server searches the knowledge database for the search data indicated by the identifier of the search data corresponding to the found search keyword.

In the data retrieval method provided by the embodiment of the present invention, the call center server can send the voice information to the voice analysis server after receiving the voice information sent by the user terminal, and the text information obtained by converting the voice information to text by the voice analysis server is sent through the call center The server forwards the text information to the text analysis server, and the text analysis server performs text analysis on the text information to identify the first word segmentation set, which contains at least one text segmentation word; the call center server retrieves from the knowledge database stored In the keyword index list of the data, search keywords matching the text word segmentation in the first word segmentation set, the keyword index list contains at least one keyword index item, and the keyword index item includes the search keyword and the The identifier of the retrieval data corresponding to the retrieval keyword, at least one retrieval data and the identifier of the retrieval data are stored in the knowledge database; the call center server retrieves the retrieval data corresponding to the retrieved retrieval keyword from the knowledge database The retrieved data indicated by the ID of the data.

Compared with the prior art when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid search errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Embodiment five

An embodiment of the present invention provides a call center server for executing the data retrieval method shown in FIG. 1, as shown in FIG. A word lookup unit 54 and a retrieval unit 55 .

The receiving unit 51 is configured to receive voice information sent by the user terminal.

The text conversion unit 52 is configured to obtain text information obtained by performing text conversion on the voice information received by the receiving unit.

The text analysis unit 53 is configured to obtain a first word segment set identified by performing text analysis on the text information obtained by the text conversion unit 52, and the first word segment set includes at least one text segment word.

A keyword search unit 54, configured to search for a search keyword that matches the text word segmentation in the first word segmentation set obtained by the text analysis unit 53 from the keyword index list of the retrieval data stored in the knowledge database, so The keyword index list includes at least one keyword index item, the keyword index item includes a search keyword and an identifier of the search data corresponding to the search keyword, and at least one search data is stored in the knowledge database and the ID of the retrieved data.

The retrieval unit 55 is configured to retrieve, from the knowledge database, the retrieval data indicated by the identifier of the retrieved data corresponding to the retrieval keyword found by the keyword searching unit 54 .

Wherein, the voice information received by the receiving unit 51 is: after the user terminal initiates a voice communication request to the call center server to establish a voice communication link between the user terminal and the call center server, the voice information received from the user terminal . Wherein, the voice information sent by the user terminal is composed of at least one voice segment whose dimension is a sentence.

Wherein, the text analysis unit 53 may perform sentence segmentation and word segmentation on the text information to obtain at least one text word segment.

Further, as shown in FIG. 6 , the call center server may further include: a reading unit 56 and a generating unit 57 .

The reading unit 56 is used to search for the text word segmentation in the first word segmentation set in the keyword index item in the keyword index list of the retrieval data stored in the knowledge database by the keyword search unit 54 Before matching the search keywords, read the first search data stored in the knowledge database, where the first search data is any search data stored in the knowledge database.

The text analysis unit 53 is further configured to obtain a second word segmentation set identified by performing text analysis on the first retrieval data read by the reading unit 56, and the second word segmentation set includes at least one text segmentation word .

A generating unit 57, configured to generate a retrieval keyword of the first retrieval data according to the text segmentation in the second segmentation set obtained by the text analysis unit 53; The identification of the first retrieval data generates one keyword index item, and generates the keyword index list including at least one keyword index item.

Further, the receiving unit 51 is also configured to search the keyword index items in the keyword index list of the retrieval data stored in the knowledge database by the keyword searching unit 54 to find Before the search keywords matched by the text word segmentation in the collection, the keyword index list sent by the text analysis server is received.

Further, in an application scenario of the embodiment of the present invention, as shown in FIG. 7 , the text analysis unit 53 includes: a text analysis module 531 and a first filtering module 532 .

The text analysis module 531 is configured to perform text analysis on the text information to identify a third word segment set, and the third word segment set includes at least one text segment word.

The first filtering module 532 is configured to filter the nonsense text participles contained in the third participle set identified by the text analysis module 531 to obtain the first participle set, and the nonsense text participles include: exclamation words and tone words.

Further, in another application scenario of the embodiment of the present invention, as shown in FIG. 8 , the text analysis unit includes: a text analysis module 531 and a second filtering module 533 .

The text analysis module 531 is configured to perform text analysis on the text information to identify a third word segment set, and the third word segment set includes at least one text segment word.

The second filtering module 533 is configured to filter the first text segmentation included in the third word segmentation set identified by the text analysis module 531 to obtain the first segmentation set, and the first text segmentation is the Text word segmentation in the third word segmentation set that does not match any search keyword in the keyword index list.

Furthermore, in another application scenario of the embodiment of the present invention, the text analysis unit includes: a text analysis module 531 , a first filter module 532 and a second filter module 533 .

In this application scenario, while the first filtering module 532 filters the meaningless text word segmentation contained in the third word segmentation set identified by the text analysis module 531, the second filtering module 533 filters the text analysis module The first text participle contained in the third participle set identified by 531; at this time, the intersection of the participle set filtered by the first filter module 532 and the participle set filtered by the second filter module 533 can be used as the first participle set.

Further, the speech information is composed of at least one speech segment whose dimension is a sentence, and the text information is composed of at least one text segment.

The receiving unit 51 is specifically configured to receive the voice segment sent by the user terminal.

The text conversion unit 52 is specifically configured to, after the receiving unit 51 receives a voice segment, send the voice segment to the voice analysis server, so that the voice analysis server converts the voice segment into the the text segment; receiving the text segment sent by the speech analysis server.

Further, the text analysis unit 53 is specifically configured to send the text information to the text analysis server, so that the text analysis server performs text analysis on the text information and identifies the first word segmentation set; The first word segmentation set sent by the text analysis server is received.

It should be noted that the call center server in the embodiment of the present invention may be a call center management server that receives calls from users from operators that can provide communication services for users. For the specific description of some functional modules in the call center server provided by the embodiment of the present invention, reference may be made to the corresponding content in the method embodiment, and details will not be repeated here in this embodiment.

The call center server provided by the embodiment of the present invention, the call center server receives the voice information sent by the user terminal, and obtains the text information obtained by performing text conversion on the voice information; the call center server obtains the first participle recognized by text analysis of the text information Set, the first word segmentation set contains at least one text segmentation word; the call center server searches the search keyword matching the text segmentation word in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, keyword index The list contains at least one keyword index item, and the keyword index item includes a retrieval keyword and an identification of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identification of the retrieval data are stored in the knowledge database; the call center The server retrieves the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword from the knowledge database.

Compared with the existing technology, when retrieving data from the knowledge database, which relies too much on the personal business experience and skills of the customer service personnel, and the accuracy and retrieval efficiency of the retrieval results are low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid retrieval errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Embodiment six

An embodiment of the present invention provides a text analysis server for executing the data retrieval method shown in FIG. 2 . As shown in FIG. 9 , the text analysis server includes: a receiving unit 61 , a text analysis unit 62 and a sending unit 63 .

The receiving unit 61 is configured to receive the text information sent by the call center server, the text information is converted by the call center server according to the voice information.

The text analysis unit 62 is configured to perform text analysis on the text information received by the receiving unit 61 to identify the first word segment set, and the first word segment set includes at least one text segment word.

The sending unit 63 is configured to send the first word segmentation set analyzed by the text analysis unit 62 to the call center server, so that the call center server obtains the information from the knowledge database according to the text word segmentation in the first word segmentation set. retrieve data from.

Wherein, the text analysis unit 62 may segment and segment the text information to identify at least one text segment.

Further, the knowledge database stores at least one retrieval data and the identification of the retrieval data.

As shown in FIG. 10 , the text analysis server may further include: a reading unit 64 and a generating unit 65 .

The reading unit 64 is configured to read the first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database.

The text analysis unit 63 is further configured to perform text analysis on the first retrieval data read by the reading unit to identify a second word segment set, and the second word segment set includes at least one text segment word.

A generating unit 65, configured to generate a retrieval keyword of the first retrieval data according to the text segmentation in the second segmentation set analyzed by the text analysis unit 63; according to the retrieval keyword of the first retrieval data and the identifier of the first retrieval data to generate one keyword index item, and generate the keyword index list including at least one keyword index item.

The sending unit 63 is further configured to send the keyword index list generated by the generating unit 65 to the call center server.

It should be noted that, for the specific description of some functional modules in the text analysis server provided by the embodiment of the present invention, reference may be made to the corresponding content in the method embodiment, and details will not be described in detail here in this embodiment.

The text analysis server provided by the embodiment of the present invention can receive the text information sent by the call center server, and the text information is converted by the call center server according to the voice information; the text analysis server performs text analysis on the text information , identifying the first word segmentation set, which contains at least one text segmentation word; the text analysis server sends the first word segmentation set to the call center server, so that the call center server according to The text word segmentation in the first word segmentation set retrieves data from the knowledge database.

Compared with the prior art, when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the text analysis server can query the call center server The sent text information is subjected to text analysis to obtain the analysis result (the first word segmentation set), and the first word segmentation set is sent to the call center server, and the call center server searches the keyword index list of the retrieval data stored in the knowledge database, and The search keywords matched by the text word segmentation in the first word segmentation set do not require the customer service personnel to identify the search keywords from the user voice information based on their personal business experience and skills, and the call center server automatically performs the recognition and comparison of the search keywords Because the customer service personnel identify the retrieval keywords based on their personal business experience and skills, it can improve the recognition efficiency of the retrieval keywords, avoid retrieval errors due to negligence or carelessness of the customer service personnel, and improve the accuracy of identifying retrieval keywords. After the search keywords are obtained efficiently and accurately, the search results can be obtained quickly and accurately.

Embodiment seven

In terms of hardware implementation, each unit in Embodiment 5 can be embedded in the processor of the call center server in the form of hardware or be independent of the processor of the call center server, and can also be stored in the memory of the call center server in the form of software, so that the processor can call and execute For the corresponding operations of the above units, the processor may be a central processing unit (CPU), a microprocessor, a single-chip microcomputer, and the like.

An embodiment of the present invention provides a call center server. As shown in FIG. 11 , the call center server includes: a receiver 71 and a processor 72 .

The receiver 71 is configured to receive voice information sent by the user terminal.

The processor 72 is configured to obtain text information obtained by performing text conversion on the voice information received by the receiver 71; obtain a first word segmentation set identified by performing text analysis on the text information, and the first word segmentation set contains at least one text word; from the keyword index list of the retrieval data stored in the knowledge database, search for a search keyword that matches the text word in the first word segmentation set, and the keyword index list contains at least one A keyword index item, the keyword index item includes a retrieval keyword and an identification of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identification of the retrieval data are stored in the knowledge database; from In the knowledge database, the retrieved data indicated by the identifier of the retrieved data corresponding to the retrieved retrieved keyword is retrieved.

Further, in an application scenario of the embodiment of the present invention, the processor 72 is also configured to search for the keyword index items in the keyword index list of the retrieved data stored in the knowledge database, which are related to the Before the retrieval keywords of the text word segmentation matching in the first word segmentation set, read the first retrieval data stored in the knowledge database, the first retrieval data is any retrieval data stored in the knowledge database; The second word segmentation set identified by the text analysis of the first search data, the second word segmentation set contains at least one text word segmentation; according to the text word segmentation in the second word segmentation set, generate the first search data Retrieve keywords; generate a keyword index item according to the search keyword of the first search data and the identifier of the first search data; generate the keyword index list including at least one keyword index item .

In another application scenario of the embodiment of the present invention, the receiver 71 is also used to include keyword index items in the keyword index list of the retrieved data stored by the processor 72 in the knowledge database , before searching for a retrieval keyword matching the text word segmentation in the first word segmentation set, receiving the keyword index list sent by the text analysis server.

In the above two application scenarios, as shown in FIG. 12 , the call center server may further include: a memory 73 .

The memory 73 is configured to store the keyword index list generated by the processor 72 ; or, the memory 73 is configured to store the keyword index list received by the receiver 71 .

Further, the processor 72 is specifically configured to perform text analysis on the text information, identify a third word segmentation set, the third word segmentation set contains at least one text segmentation word; filter the third word segmentation set contains The nonsense text word segmentation to obtain the first word segmentation set, the nonsense text word segmentation includes: exclamation words and modal particles.

Further, the processor 72 is specifically configured to perform text analysis on the text information, identify a third word segmentation set, the third word segmentation set contains at least one text segmentation word; filter the third word segmentation set contains The first text segmentation to obtain the first word segmentation set, the first text segmentation is the text segmentation in the third word segmentation set that does not match any search keyword in the keyword index list.

Further, the speech information is composed of at least one speech segment whose dimension is a sentence, and the text information is composed of at least one text segment.

The receiver 71 is further configured to receive the voice segment sent by the user terminal.

As shown in FIG. 13 , the call center server may further include: a sender 74 .

The sender 74 is configured to send the voice segment to a voice analysis server after the receiver 71 receives the voice segment, so that the voice analysis server converts the voice segment into the text segment.

The receiver 71 is further configured to receive the text segment sent by the speech analysis server.

Further, the sender 74 is further configured to send the text information to the text analysis server, so that the text analysis server performs text analysis on the text information and identifies the first word segmentation set.

The receiver 71 is further configured to receive the first word segmentation set sent by the text analysis server.

In the embodiment of the present invention, the receiver 71, the processor 72, the memory 73, and the transmitter 74 are connected through a bus and complete mutual communication.

Wherein, the bus may be an Industry Standard Architecture (Industry Standard Architecture, ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA) bus. The bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in FIGS. 11-13 , but it does not mean that there is only one bus or one type of bus.

The processor 72 may be a central processing unit (Central Processing Unit, CPU), or a combination of a CPU, a digital signal processor (Digital Signal Processor, DSP) and a control chip (such as a baseband chip) in the communication unit. In this embodiment of the present invention, the CPU may be a single computing core, or may include multiple computing cores.

The receiver 71 and the transmitter 74 may be devices that send and receive wireless signals through antennas, or other devices that provide signal sending and receiving interfaces.

It should be noted that, for the specific description of some functional modules in the call center server provided by the embodiment of the present invention, reference may be made to the corresponding content in the method embodiment, and details will not be repeated here in this embodiment.

The call center server provided by the embodiment of the present invention can receive the voice information sent by the user terminal, and obtain the text information obtained by performing text conversion on the voice information; the call center server obtains the first word segmentation set identified by text analysis on the text information, The first word segmentation set contains at least one text segmentation word; the call center server searches the retrieval keyword matching the text segmentation word in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, and in the keyword index list Including at least one keyword index item, the keyword index item includes a retrieval keyword and an identifier of the retrieval data corresponding to the retrieval keyword, and at least one retrieval data and the identifier of the retrieval data are stored in the knowledge database; In the knowledge database, the retrieved data indicated by the identifier of the retrieved data corresponding to the retrieved keyword is retrieved.

Compared with the prior art when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the call center server receives the user terminal After the voice information is sent, the analysis result (the first word segmentation set) can be obtained by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, search for the text word segmentation in the first word segmentation set Matched search keywords do not require customer service personnel to identify search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes search keywords compared to customer service personnel based on their personal business Recognition of search keywords based on experience and skills can improve the recognition efficiency of search keywords, avoid search errors due to negligence or carelessness of customer service personnel, improve the accuracy of recognition of search keywords, and obtain search keywords efficiently and accurately. After keywords, you can quickly and accurately obtain the search results.

Embodiment eight

In terms of hardware implementation, each unit in Embodiment 6 can be embedded in or independent of the processor of the text analysis server in the form of hardware, and can also be stored in the memory of the text analysis server in the form of software, so that the processor can call and execute For the corresponding operations of the above units, the processor may be a central processing unit (CPU), a microprocessor, a single-chip microcomputer, and the like.

An embodiment of the present invention provides a text analysis server, as shown in FIG. 14 , including: a receiver 81 , a processor 82 and a sender 83 .

The receiver 81 is configured to receive the text information sent by the call center server, and the text information is converted by the call center server according to the voice information.

The processor 82 is configured to perform text analysis on the text information received by the receiver 81 to identify the first word segment set, and the first word segment set includes at least one text segment word.

The sender 83 is configured to send the first word segmentation set identified by the processor 82 to the call center server, so that the call center server can learn from the knowledge database according to the text word segmentation in the first word segmentation set. Retrieve data.

Further, the knowledge database stores at least one retrieval data and the identification of the retrieval data.

The processor 82 is also configured to read the first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database; Text analysis, identifying a second word segmentation set, which contains at least one text segmentation; according to the text segmentation in the second word segmentation set, generating a retrieval keyword for the first retrieval data; according to the The retrieval keyword of the first retrieval data and the identifier of the first retrieval data generate a keyword index item; generate the keyword index list including at least one keyword index item.

The sender 83 is further configured to send the keyword index list generated by the processor 82 to the call center server.

In the embodiment of the present invention, the receiver 81, the processor 82, and the transmitter 83 are connected through a bus to complete mutual communication.

Wherein, the bus may be an Industry Standard Architecture (Industry Standard Architecture, ISA) bus, a Peripheral Component Interconnect (PCI) bus, or an Extended Industry Standard Architecture (Extended Industry Standard Architecture, EISA) bus. The bus can be divided into address bus, data bus, control bus and so on. For ease of representation, only one thick line is used in FIG. 14 , but it does not mean that there is only one bus or one type of bus.

The processor 82 may be a central processing unit (Central Processing Unit, CPU), or a combination of a CPU, a digital signal processor (Digital Signal Processor, DSP) and a control chip (such as a baseband chip) in the communication unit. In this embodiment of the present invention, the CPU may be a single computing core, or may include multiple computing cores.

The receiver 81 and the transmitter 82 may be devices that send and receive wireless signals through antennas, or other devices that provide signal sending and receiving interfaces.

It should be noted that, for the specific description of some functional modules in the text analysis server provided by the embodiment of the present invention, reference may be made to the corresponding content in the method embodiment, and details will not be described in detail here in this embodiment.

The text analysis server provided by the embodiment of the present invention can receive the text information sent by the call center server, and the text information is converted by the call center server according to the voice information; the text analysis server performs text analysis on the text information , identifying the first word segmentation set, which contains at least one text segmentation word; the text analysis server sends the first word segmentation set to the call center server, so that the call center server according to The text word segmentation in the first word segmentation set retrieves data from the knowledge database.

Compared with the prior art, when retrieving data from the knowledge database, relying too much on the personal business experience and skills of the customer service personnel, the accuracy and retrieval efficiency of the retrieval results are relatively low, through this scheme, the text analysis server can query the call center server The sent text information is subjected to text analysis to obtain the analysis result (the first word segmentation set), and the first word segmentation set is sent to the call center server, and the call center server searches the keyword index list of the retrieval data stored in the knowledge database, and The search keywords matched by the text word segmentation in the first word segmentation set do not require the customer service personnel to identify the search keywords from the user voice information based on their personal business experience and skills, and the call center server automatically performs the recognition and comparison of the search keywords Because the customer service personnel identify the retrieval keywords based on their personal business experience and skills, it can improve the recognition efficiency of the retrieval keywords, avoid retrieval errors due to negligence or carelessness of the customer service personnel, and improve the accuracy of identifying retrieval keywords. After the search keywords are obtained efficiently and accurately, the search results can be obtained quickly and accurately.

It should be noted that, in the first application scenario of the embodiment of the present invention, the call center server may be integrated with a text conversion module capable of performing text conversion on voice information and a text conversion module capable of performing text conversion on voice information. The text analysis module for performing text analysis on the obtained text information, that is, the function of performing text conversion on the voice information within the call center server, and then performing text analysis on the text information obtained through the text conversion.

In the second application scenario of the embodiment of the present invention, the call center server needs to send voice information to the voice analysis server, and the voice analysis server implements text conversion of the voice information to obtain text information, and then receives the text information sent by the voice analysis server ; The call center server needs to send text information to the text analysis server, and the text analysis server implements text analysis on the text information to obtain the analysis result, and then receives the analysis result sent by the text analysis server.

Corresponding to the second application scenario, the embodiment of the present invention provides a data retrieval system. As shown in FIG. 15 , the data retrieval system includes: a call center server 91 , a voice analysis server 92 , a text analysis server 93 and a knowledge database 94 .

The call center server 91 is configured to receive the voice information sent by the user terminal, and send the voice information to the voice analysis server 92, the voice information is composed of at least one voice segment with a sentence as the dimension.

The speech analysis server 92 is configured to convert the speech segment into the text segment, and send text information to the call center server 91, the text information is composed of at least one text segment.

The call center server 91 is further configured to send the text information to the text analysis server 93 .

The text analysis server 93 is configured to perform text analysis on the text information to identify a first word segment set, and the first word segment set includes at least one text segment word.

The call center server 91 is further configured to search for a search keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, in the keyword index list Contains at least one keyword index item, the keyword index item includes a search keyword and an identifier of the search data corresponding to the search keyword, and the knowledge database stores at least one search data and an identifier of the search data An identifier: from the knowledge database, retrieve the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieval keyword.

Further, the text analysis server 93 is further configured to send the keyword index list to the call center server 91 .

Further, the text analysis server 93 is also used to read the first retrieval data stored in the knowledge database before sending the keyword index list to the call center server 91, the first retrieval data For any retrieval data stored in the knowledge database; performing text analysis on the first retrieval data to identify a second word segmentation set, which contains at least one text segmentation word; according to the second word segmentation The text word segmentation in the set, generate the retrieval keyword of the first retrieval data; generate a keyword index item according to the retrieval keyword of the first retrieval data and the identification of the first retrieval data; generate a keyword index item containing at least The keyword index list of one of the keyword index items.

Compared with the data retrieval system provided by the embodiment of the present invention, when retrieving data from the knowledge database in the prior art, it relies too much on the personal business experience and skills of the customer service personnel, and the accuracy and retrieval efficiency of the retrieval results are relatively low. Solution, after receiving the voice information sent by the user terminal, the call center server can obtain the analysis result (the first word segmentation set) by performing text analysis on the voice information, and then from the keyword index list of the retrieval data stored in the knowledge database, Find search keywords that match the text word segmentation in the first word segmentation set, without the need for customer service personnel to identify the search keywords from user voice information based on their personal business experience and skills, and the call center server automatically recognizes the search keywords Compared with the identification of search keywords by customer service personnel based on their personal business experience and skills, it can improve the identification efficiency of search keywords, avoid retrieval errors due to negligence or carelessness of customer service personnel, and improve the accuracy of identifying search keywords After obtaining the retrieval keywords efficiently and accurately, the retrieval results can be obtained quickly and accurately.

Through the description of the above embodiments, those skilled in the art can clearly understand that for the convenience and brevity of the description, only the division of the above-mentioned functional modules is used as an example for illustration. In practical applications, the above-mentioned functions can be allocated according to needs It is completed by different functional modules, that is, the internal structure of the device is divided into different functional modules to complete all or part of the functions described above. For the specific working process of the above-described system, device, and unit, reference may be made to the corresponding process in the foregoing method embodiments, and details are not repeated here.

In the several embodiments provided in this application, it should be understood that the disclosed system, device and method can be implemented in other ways. For example, the device embodiments described above are only illustrative. For example, the division of the modules or units is only a logical function division. In actual implementation, there may be other division methods. For example, multiple units or components can be Incorporation may either be integrated into another system, or some features may be omitted, or not implemented. In another point, the mutual coupling or direct coupling or communication connection shown or discussed may be through some interfaces, and the indirect coupling or communication connection of devices or units may be in electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place, or may be distributed to multiple network units. Part or all of the units can be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, each unit may exist separately physically, or two or more units may be integrated into one unit. The above-mentioned integrated units can be implemented in the form of hardware or in the form of software functional units.

If the integrated unit is realized in the form of a software function unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the essence of the technical solution of the present invention or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product, and the computer software product is stored in a storage medium , including several instructions to make a computer device (which may be a personal computer, a server, or a network device, etc.) or a processor (processor) execute all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-OnlyMemory), random access memory (RAM, RandomAccessMemory), magnetic disk or optical disk and other media that can store program codes.

The above is only a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Anyone skilled in the art can easily think of changes or substitutions within the technical scope disclosed in the present invention. Should be covered within the protection scope of the present invention. Therefore, the protection scope of the present invention should be determined by the protection scope of the claims.

Claims (18)

1. A data retrieval method, characterized in that, comprising: The call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text; The call center server acquires a first word segmentation set identified by performing text analysis on the text information, and the first word segmentation set includes at least one text segmentation word; The call center server searches for a retrieval keyword that matches the text word segmentation in the first word segmentation set from the keyword index list of the retrieval data stored in the knowledge database, and the keyword index list contains at least one keyword An index item, the keyword index item includes a search keyword and an identifier of the search data corresponding to the search keyword, and at least one search data and the identifier of the search data are stored in the knowledge database; The call center server retrieves, from the knowledge database, the retrieved data indicated by the identifier of the retrieved data corresponding to the found retrieved keyword. 2. method according to claim 1, is characterized in that, in the keyword index item in the keyword index list in the retrieval data that described call center server stores from knowledge database, search and described first word segmentation set Before the search keyword matched by the text word segmentation, the method also includes: The call center server reads the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database; The call center server obtains a second word segmentation set identified by performing text analysis on the first retrieval data, and the second word segmentation set includes at least one text segmentation word; The call center server generates the retrieval keywords of the first retrieval data according to the text segmentation in the second segmentation set; The call center server generates a keyword index item according to the retrieval keyword of the first retrieval data and the identifier of the first retrieval data; The call center server generates the keyword index list including at least one keyword index item. 3. The method according to claim 1, characterized in that, in the keyword index item in the keyword index list of the retrieved data stored by the call center server from the knowledge database, search for the keyword index item related to the first Before the retrieval keyword matched by the text word segmentation in the word segmentation set, the method also includes: The call center server receives the keyword index list sent by the text analysis server. 4. according to the method described in any one in claim 1-3, it is characterized in that, described call center server obtains the first participle set that text analysis is identified to described text information, comprises: The call center server performs text analysis on the text information, and recognizes a third participle set, which contains at least one text participle; The call center server filters the meaningless text participles included in the third participle set to obtain the first participle set, and the meaningless text participle includes: exclamation words and modal particles. 5. according to the method described in any one in claim 1-3, it is characterized in that, described call center server obtains the first participle set that the text analysis of described text information is identified, comprises: The call center server performs text analysis on the text information, and recognizes a third participle set, which contains at least one text participle; The call center server filters the first text segmentation included in the third word segmentation set to obtain the first text segmentation set, and the first text segmentation is the combination of the third word segmentation set and the keyword index list A text segment that does not match any of the search keywords. 6. The method according to claim 1 or 3, wherein the speech information is composed of at least one speech segment taking a sentence as a dimension, and the text information is composed of at least one text segment; The call center server receives the voice information sent by the user terminal, and obtains the text information obtained by converting the voice information into text, including: The call center server receives the voice segment sent by the user terminal; The call center server sends the voice segment to the voice analysis server after receiving one of the voice segments, so that the voice analysis server converts the voice segment into the text segment; The call center server receives the text segment sent by the speech analysis server. 7. The method according to claim 1 or 3, wherein the call center server acquires the first word segmentation set identified by text analysis of the text information, including: The call center server sends the text information to the text analysis server, so that the text analysis server performs text analysis on the text information and identifies the first word segmentation set; The call center server receives the first word segmentation set sent by the text analysis server. 8. A data retrieval method, characterized in that, comprising: The text analysis server receives the text information sent by the call center server, and the text information is converted by the call center server according to the voice information; The text analysis server performs text analysis on the text information, and identifies the first word segment set, and the first word segment set contains at least one text segment word; The text analysis server sends the first word segmentation set to the call center server, so that the call center server retrieves data from the knowledge database according to the text segmentation words in the first word segmentation set. 9. The method according to claim 8, wherein at least one retrieval data and an identifier of the retrieval data are stored in the knowledge database; The method also includes: The text analysis server reads the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database; The text analysis server performs text analysis on the first retrieval data, and identifies a second word segmentation set, and the second word segmentation set contains at least one text segmentation word; The text analysis server generates the retrieval keywords of the first retrieval data according to the text segmentation in the second segmentation set; The text analysis server generates one keyword index item according to the retrieval keyword of the first retrieval data and the identifier of the first retrieval data; The text server generates the keyword index list including at least one keyword index item; The text analysis server sends the keyword index list to the call center server. 10. A call center server, characterized in that, comprising: a receiving unit, configured to receive voice information sent by the user terminal; a text conversion unit, configured to obtain text information obtained by performing text conversion on the voice information received by the receiving unit; a text analysis unit, configured to obtain a first word segmentation set identified by performing text analysis on the text information obtained by the text conversion unit, and the first word segmentation set includes at least one text segment; A keyword search unit, configured to search for a search keyword that matches the text word segmentation in the first word segmentation set obtained by the text analysis unit from the keyword index list of the retrieval data stored in the knowledge database, the keyword The word index list contains at least one keyword index item, and the keyword index item includes a search keyword and an identification of the search data corresponding to the search keyword, and at least one search data and the search data are stored in the knowledge database. identification of the retrieved data; The retrieval unit is configured to retrieve, from the knowledge database, the retrieval data indicated by the identifier of the retrieved data corresponding to the retrieval keyword found by the keyword searching unit. 11. The call center server according to claim 10, further comprising: A reading unit, configured to search for, among the keyword index items in the keyword index list of the retrieval data stored in the knowledge database by the keyword search unit, the words that match the text word segmentation in the first word segmentation set Before retrieving keywords, read the first retrieval data stored in the knowledge database, and the first retrieval data is any retrieval data stored in the knowledge database; The text analysis unit is further configured to obtain a second word segmentation set identified by performing text analysis on the first retrieval data read by the reading unit, and the second word segmentation set includes at least one text segmentation word; A generation unit, configured to generate a retrieval keyword of the first retrieval data according to the text word segmentation in the second word segmentation set obtained by the text analysis unit; according to the retrieval keyword of the first retrieval data and the The identification of the first retrieval data generates one keyword index item, and generates the keyword index list including at least one keyword index item. 12. The call center server according to claim 10, characterized in that, the receiving unit is also used for key words in the keyword index list of the retrieved data stored in the knowledge database by the keyword search unit In the word index item, before searching for the search keywords matching the text word segments in the first word segment set, the keyword index list sent by the text analysis server is received. 13. The call center server according to any one of claims 10-12, wherein the text analysis unit includes: A text analysis module, configured to perform text analysis on the text information, and identify a third word segment set, the third word segment set includes at least one text segment word; The first filtering module is used to filter the nonsense text participles contained in the third participle set identified by the text analysis module to obtain the first participle set, and the nonsense text participles include: exclamation words and tone word. 14. The call center server according to any one of claims 10-12, wherein the text analysis unit includes: A text analysis module, configured to perform text analysis on the text information, and identify a third word segment set, the third word segment set includes at least one text segment word; The second filtering module is configured to filter the first text segmentation contained in the third segmentation set identified by the text analysis module to obtain the first segmentation set, and the first text segmentation is the third segmentation Text word segmentation in the collection that does not match any search keyword in the keyword index list. 15. The call center server according to claim 10 or 12, wherein the voice information is composed of at least one voice segment taking a sentence as a dimension, and the text information is composed of at least one text segment; The receiving unit is specifically configured to receive the voice segment sent by the user terminal; The text conversion unit is specifically configured to, after the receiving unit receives a voice segment, send the voice segment to a voice analysis server, so that the voice analysis server converts the voice segment into the text Fragment: receiving the text fragment sent by the speech analysis server. 16. The call center server according to claim 10 or 12, wherein the text analysis unit is specifically configured to send the text information to the text analysis server, so that the text analysis server can analyze the Performing text analysis on the text information to identify the first word segmentation set; receiving the first word segmentation set sent by the text analysis server. 17. A text analysis server, characterized in that, comprising: a receiving unit, configured to receive text information sent by the call center server, the text information being converted by the call center server according to the voice information; A text analysis unit, configured to perform text analysis on the text information received by the receiving unit, and identify the first word segment set, the first word segment set includes at least one text segment word; A sending unit, configured to send the first word segmentation set analyzed by the text analysis unit to the call center server, so that the call center server retrieves from the knowledge database according to the text word segmentation in the first word segmentation set data. 18. The text analysis server according to claim 17, wherein at least one retrieval data and an identifier of the retrieval data are stored in the knowledge database; The text analysis server also includes: A reading unit, configured to read first retrieval data stored in the knowledge database, where the first retrieval data is any retrieval data stored in the knowledge database; The text analysis unit is further configured to perform text analysis on the first retrieval data read by the reading unit, and identify a second word segment set, the second word segment set includes at least one text segment word; A generating unit, configured to generate a retrieval keyword of the first retrieval data according to the text segmentation in the second segmentation set analyzed by the text analysis unit; according to the retrieval keyword of the first retrieval data and the The identification of the first retrieval data generates one of the keyword index items, and generates the keyword index list including at least one of the keyword index items; The sending unit is further configured to send the keyword index list generated by the generating unit to the call center server.