WO2018214859A1 - Method and device for analyzing man-machine dialogue data - Google Patents

Abstract

The present disclosure provides a method and a device for analyzing man-machine dialogue data and is applied to a data analysis apparatus. The method comprises: determining data to be analyzed; providing a data analysis interactive interface for receiving a query instruction input by a user at the data analysis interactive interface, and displaying the query instruction in the form of interactive dialogue; statistically analyzing, according to the query instruction, the data to be analyzed to obtain corresponding statistical data; determining, according to the query instruction, a data display manner, and generating, according to the data display manner, a visualized analysis result of the statistical data; and displaying the visualized analysis result in the form of interactive dialogue on the data analysis interactive interface. Hence, by carrying out interaction in the form of a human-machine dialogue, the data analysis operations are more concise and intuitive, the interaction manner is more humanized, and the difficulty threshold of data analysis functions is reduced.

Description

Man-machine dialogue data analysis method and device

Cross-reference to related applications

The present disclosure claims priority to Chinese Patent Application No. 2017103639701, entitled "Human-Machine Dialogue Data Analysis Method and Apparatus", filed on May 22, 2017, the entire contents of which are incorporated herein by reference. .

Technical field

The present disclosure relates to the field of data analysis technologies, and in particular, to a human-machine dialog data analysis method and apparatus.

Background technique

With the development of digital information technology, the analysis of data processing has become more and more important, and data analysis plays a huge role in all walks of life. At the same time, the way of data analysis is becoming more and more intelligent, and the data processing device can analyze the specified data according to the analysis command input by the user and feed back the analysis result. However, there is a long-standing problem with the prior art data analysis methods, that is, professionals need to perform data analysis, and the recruitment and training of professionals consumes a large amount of time and labor costs of the enterprise, which requires special use in specialized database software. The database query language performs data query analysis, and the operations of confirming the analysis target and inputting the analysis content are cumbersome, the interaction mode is not human, and the operability is not strong. As ordinary users who do not have professional data analysis knowledge or do not understand database operations are increasingly demanding data analysis, how to provide a simple and effective data analysis interaction method has long plagued the technical personnel in the field, which hinders data analysis. Wide application of methods.

Summary of the invention

In order to overcome at least one of the deficiencies in the prior art, an object of the present disclosure is to provide a human-machine dialog data analysis method, which is applied to a data analysis system, and the method includes:

Determining the data to be analyzed;

Providing a data analysis interaction interface, receiving a query instruction input by the user on the data analysis interaction interface, and displaying the query instruction in a dialog interaction manner, wherein the query instruction includes the query content;

Performing statistical analysis on the data to be analyzed according to the query instruction, and obtaining corresponding statistical data;

Determining, according to the query instruction, a data display manner, and generating a visual analysis result of the statistical data according to the data display manner;

The visual analysis result is displayed in the form of a dialog interaction at the data analysis interaction interface.

Optionally, the step of determining data to be analyzed includes:

Providing a data selection interface;

In response to the user's operation on the data selection interface, the selected data is taken as the data to be analyzed.

Optionally, the receiving the query instruction input by the user in the data analysis interaction interface, and displaying the query instruction in the form of a dialog interaction, including any one of the following:

Receiving and displaying text information input by the user, and displaying the query instruction in the form of a dialog interaction; or

The voice information input by the user through the voice input device is received, the voice information is converted into a text command, and the text command is displayed in the form of a dialog interaction.

Optionally, the step of providing a data analysis interaction interface, receiving a query instruction input by the user on the data analysis interaction interface, and displaying the query instruction in the form of a dialog interaction, includes:

Providing a data analysis interaction interface by using a third-party communication software configured on the user terminal, and receiving a query instruction sent by the user terminal through the third-party communication software;

The step of displaying the visual analysis result in the form of a dialog interaction in the data analysis interaction interface includes:

Sending the generated visual analysis result to the user terminal, so that the user terminal displays the visual analysis result in the form of a dialog interaction through a data analysis interaction interface provided by the third-party communication software.

Optionally, the query instruction includes a statement in a natural language form; and the step of performing statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, including:

Parsing the statement in the natural language form to obtain a database query analysis statement;

Performing statistical analysis on the data to be analyzed according to the database query analysis statement to obtain corresponding statistical data.

Optionally, the step of determining data to be analyzed includes:

Determining the data to be analyzed, and determining a target technical field to which the data to be analyzed belongs;

The step of parsing the statement in the natural language form to obtain a database query analysis statement includes:

Determining, by the corresponding language analysis model, the query instruction of the natural language form according to the target technical field to obtain the database query analysis statement.

Optionally, the step of determining the target technical field of the data to be analyzed includes:

Responding to user operations, using the technical field selected by the user as the target technical field; or

Extracting a file name of the data to be analyzed and a name of each data item in the data to be analyzed, performing statistics on the keyword in the file name and the data item name, and determining the data to be analyzed according to the statistical result. Target technology area.

Optionally, the step of determining a data presentation manner according to the query instruction, and generating a visual analysis result of the statistical data according to the data display manner, includes:

Determining whether the specified data display mode is included in the query instruction;

When the specified data display mode is not included in the query instruction, the visual analysis result of the statistical data is generated according to a default data display manner;

When the specified data display mode is included in the query instruction, the visual analysis result of the statistical data is generated according to the specified data display manner.

Optionally, the method further includes:

Receiving and displaying a default data display mode configuration instruction input by the user in the data analysis interaction interface, where the default data display mode configuration instruction includes a default data display mode specified by the user, and configuring a visual analysis result according to the default data display mode configuration instruction. The default data display method.

Optionally, the data display manner includes a graphic display manner; the graphic display manner includes one or more of a graphic style, a color quantity, a color type, and a graphic size; and the graphic style includes: a line drawing, a histogram, Dashboard, pie, radar, heat, funnel, or scatter plots.

Optionally, the method further includes:

Receiving and displaying a display mode update instruction input by the user on the data analysis interaction interface, where the display mode update instruction includes a data display mode;

And re-generating the visual analysis result of the statistical data according to the data display manner in the display mode update instruction;

The newly generated visual analysis result is displayed on the data analysis interactive interface.

Optionally, the method further includes:

Receiving and displaying a statistical range update instruction input by the user on the data analysis interaction interface;

The statistical range corresponding to the current visual analysis result is re-determined or modified according to the statistical range update instruction, and a new visual analysis result is generated.

Optionally, the method further includes:

Receiving, by the user, the selection of the visual analysis result on the data analysis interaction interface, and storing or transmitting the selected visual analysis result.

Optionally, the method further includes:

Recording a query instruction input by the user on the data analysis interaction interface, and the data analysis system generates a visual analysis result in response to the query instruction, and generates an analysis interaction record;

The analytical interaction record is stored or transmitted in response to user operations.

Optionally, before the step of performing statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, the method further includes:

Performing a hash calculation on the file identifier of the data to be analyzed and the keyword in the query instruction to obtain a query hash value;

Find if the visual analysis result associated with the query hash value has been generated;

If the visual analysis result associated with the query hash value is queried, performing the step of displaying the visual analysis result in the form of a dialog interaction in the data analysis interaction interface;

If the result of the visual analysis associated with the query hash value is not queried, performing a statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, and obtaining the corresponding statistical data according to the statistical data After visualizing the analysis result, the visual analysis result is stored, and the query hash value is associated with the obtained visual analysis result.

Optionally, the method further includes:

And displaying, by the first preset duration, the visual analysis result that the storage time exceeds the second preset duration, wherein the first preset duration is greater than the second preset duration.

Another object of the present invention is to provide a device for analyzing human-machine dialog data, which is applied to a data analysis system, and the device includes:

a file determination module configured to determine data to be analyzed;

The instruction acquisition module is configured to provide a data analysis interaction interface, and receive and display a query instruction input by the user on the data analysis interaction interface, where the query instruction includes the query content;

The statistical analysis module is configured to perform statistical analysis on the data to be analyzed according to the query instruction, to obtain corresponding statistical data;

a result generating module, configured to determine a data display manner according to the query instruction, and generate a visual analysis result of the statistical data according to the data display manner;

The result display module is configured to display the visual analysis result on the data analysis interaction interface.

Optionally, the file determining module is specifically configured to provide a data selection interface; in response to the operation of the user in the data selection interface, the selected data is used as data to be analyzed.

Optionally, the instruction acquiring module is specifically configured to receive and display text information input by the user, and display the query instruction in a dialog interaction manner; or receive voice information input by the user through the voice input device, and the voice information is Convert to a text instruction and display the text instruction in the form of a dialog interaction.

Optionally, the instruction obtaining module is specifically configured to provide a data analysis interaction interface by using a third-party communication software configured on the user terminal, and receive a query instruction sent by the user terminal by using the third-party communication software;

The result display module is specifically configured to send the generated visual analysis result to the user terminal, so that the user terminal displays the visual analysis result in the form of a dialog interaction through a data analysis interaction interface provided by the third-party communication software. .

It is another object of the present disclosure to provide a readable storage medium storing machine executable instructions that, when executed by a processor, cause the processor to implement the present disclosure Human-machine dialogue data analysis method.

Compared with the prior art, the present disclosure has at least the following beneficial effects:

The human-machine dialog data analysis method and apparatus provided by the embodiments of the present disclosure receive a query instruction of the user in the interface data by providing a friendly data analysis interaction interface, and perform data analysis according to the query instruction, in the form of human-machine dialogue. The analysis result is displayed on the analysis interaction interface. In this way, interaction through human-machine dialogue makes the operation of data analysis more concise and intuitive, and the interaction mode is more user-friendly, which reduces the threshold for using data analysis functions.

DRAWINGS

In order to more clearly illustrate the technical solutions of the embodiments of the present disclosure, the drawings to be used in the embodiments will be briefly described below. It should be understood that the following drawings show only certain embodiments of the present disclosure, and thus It should be seen as a limitation on the scope, and those skilled in the art can obtain other related drawings according to these drawings without any creative work.

FIG. 1 is a schematic diagram of interaction between a data analysis device and a user terminal according to an embodiment of the present disclosure;

2 is a schematic structural diagram of hardware of a data analysis device according to an embodiment of the present disclosure;

FIG. 3 is a schematic flowchart of steps of a human-machine dialog data analysis method according to an embodiment of the present disclosure;

FIG. 4 is a schematic diagram of a file selection interface according to an embodiment of the present disclosure;

FIG. 5 is a schematic diagram of a data analysis interaction interface according to an embodiment of the present disclosure;

FIG. 6 is a schematic diagram of a human-machine dialog data analyzing apparatus according to an embodiment of the present disclosure.

Icon: 100-data analysis device; 110-human-machine dialog data analysis device; 111-file determination module; 112-instruction acquisition module; 113-statistical analysis module; 114-result generation module; 115-result display module; Terminal; 300-network.

detailed description

The technical solutions in the embodiments of the present disclosure will be clearly and completely described in conjunction with the drawings in the embodiments of the present disclosure. It is a partial embodiment of the present disclosure, and not all of the embodiments. The components of the disclosed embodiments, which are generally described and illustrated in the figures herein, can be arranged and designed in various different configurations.

The detailed description of the embodiments of the present disclosure, which is set forth in the claims All other embodiments obtained by a person of ordinary skill in the art based on the embodiments of the present disclosure without departing from the inventive scope are the scope of the disclosure.

It should be noted that similar reference numerals and letters indicate similar items in the following figures, and therefore, once an item is defined in a drawing, it is not necessary to further define and explain it in the subsequent drawings.

In the description of the present disclosure, it should be further noted that the terms "set", "install", "connected", and "connected" are to be understood broadly, and may be fixed connections, for example, unless otherwise specifically defined and defined. It can also be a detachable connection, or an integral connection; it can be a mechanical connection or an electrical connection; it can be directly connected or indirectly connected through an intermediate medium, and can be internal communication between the two elements. The specific meanings of the above terms in the present disclosure can be understood in the specific circumstances by those skilled in the art.

The human-machine dialog data analysis method provided in this embodiment is applied to a data analysis device. Referring to FIG. 1 , in an example, the data analysis device 100 may be connected to the user terminal 200 through a network 300. The user terminal 200 receives a user input query analysis content and sends the data to the data analysis device 100 for data analysis. And receiving the analysis result fed back by the data analysis device 100. In another example, the data analysis device 100 can also independently receive the query analysis content input by the user, and analyze the result to feed back the analysis result to the user.

Please refer to FIG. 2. FIG. 2 is a block diagram of the data analysis device 100 shown in FIG. 1. The data analysis device 100 includes a human-machine dialog data analysis device 110, a memory 120, and a processor 130.

The components of the memory 120 and the processor 130 are electrically connected directly or indirectly to each other to implement data transmission or interaction. For example, the components can be electrically connected to one another via one or more communication buses or signal lines. The human-machine dialog data analyzing device 110 includes at least one software function that can be stored in the memory 120 or solidified in an operating system (OS) of the data analyzing device 100 in the form of software or firmware. Module. The processor 130 is configured to execute an executable module stored in the memory 120, for example, a software function module, a computer program, and the like included in the human-machine dialog data analyzing device 110.

The memory 120 may be, but not limited to, a random access memory (RAM), a read only memory (ROM), and a programmable read-only memory (PROM). Erasable Programmable Read-Only Memory (EPROM), Electric Erasable Programmable Read-Only Memory (EEPROM), and the like. The memory 120 is configured to store a program, and the processor 130 executes the program after receiving an execution instruction.

Please refer to FIG. 2. FIG. 2 is a flowchart of a human-machine dialog data analysis method applied to the data analysis device 100 shown in FIG. 1, and the steps including the steps will be described in detail below.

In step S110, the data to be analyzed is determined.

Referring to FIG. 3, in the embodiment, the data analysis device 100 can provide a data selection interface.

In one example, the data analysis device 100 can provide a web page, the data molecule device 100 further including a communication unit configured to perform data interaction with the user terminal 200, the data analysis device 100 through the communication The unit receives the access request sent by the user terminal 200, and sends the web page to the user terminal 200, so that the user terminal 200 loads the web page to display the data selection interface; or the data analysis device 100 can also A special APP installation package is provided, so that the user terminal 200 can obtain a special APP by downloading and installing the APP installation package, and display the data selection interface after launching the APP.

In another example, the data analysis device 100 can include or be coupled to at least one display device through which the data analysis device 100 can display the data selection interface.

The data analysis device 100 can display at least one data file on the data selection interface. For example, a plurality of data files uploaded by the user may be displayed; or a plurality of data files may be determined and displayed in the data analysis device 100 in response to user operations; or data interaction with other devices may be performed in response to user operations to obtain from other devices. At least one data file specified by the user. The data file may include, but is not limited to, an EXCEL data file, a relational database table, and the like. The data file may be arranged to display the plurality of data files by means of a chart and/or a text description, and at the same time, information such as the number of rows of data in each data file and the date of creation of the data file may be realized. In this way, the user can visually see the data file on the data selection interface.

The processor 130 of the data analysis device 100 can then use the selected data as data to be analyzed in response to the user's operation at the data selection interface. The data analysis device 100 may use one or more data selected by the user as data to be analyzed.

For example, the data analysis device 100 may respond to a user's click on the data file on the file selection interface, and use the data file clicked by the user as the data to be analyzed. The data to be analyzed may also be determined for an instruction input by text or voice.

Step S120, providing a data analysis interaction interface, receiving a query instruction input by the user on the data analysis interaction interface, and displaying the query instruction in the form of a dialog interaction, wherein the query instruction includes the query content.

Referring to FIG. 4, after the user determines the data to be analyzed, the data analysis interface displays a data analysis interaction interface. The data analysis device 100 may provide the data analysis interaction interface in the same or similar manner to provide the data selection interface. For details, refer to the description of the data selection interface in step S110, and details are not described herein again.

The processor 130 of the data analysis device 100 can receive a query instruction input by the user at the data analysis interaction interface. The query instructions may include the need to query content, such as "showing revenue trends for each month of the most recent year." The query instruction may also include a combination of the query content and the specified data presentation manner, such as "using a line chart to show the revenue trend of each month in the most recent year."

The manner of displaying the query instruction in the form of a dialog interaction, including any one of the following:

Receiving and displaying text information input by the user, and displaying the query instruction in the form of a dialog interaction; or

The voice information input by the user through the voice input device is received, the voice information is converted into a text command, and the text command is displayed in the form of a dialog interaction.

For example, a text input box may be displayed on the data analysis interaction interface, and the user may input text information in a text input box, and the data analysis device 100 displays the form in a dialogue interaction in the data analysis interaction interface. text information. A voice input button may also be displayed on the data analysis interface, the user may input voice information by clicking a voice input button, and the data analysis device 100 converts the voice information into text information and displays the voice information in the form of a dialog interaction. Data analysis on the interactive interface.

Step S130: Perform statistical analysis on the data to be analyzed according to the query instruction, and obtain corresponding statistical data.

Optionally, the query instruction comprises a statement in the form of a natural language. The data analysis device 100 parses the received statement in the natural language form to obtain a structured database query analysis statement, for example, a SQL (Structured Query Language) query statement. Then, according to the database query analysis statement, statistical analysis is performed on the data to be analyzed, and corresponding statistical data is obtained. In this way, users who do not understand the professional database query statement can also perform data analysis through the natural language to issue query instructions, which is convenient for the user to use.

In this embodiment, the processor 130 of the data analysis device 100 may obtain the database query statement by analyzing the statement in the natural language form by using a general semantic analysis algorithm, for example, using a semantic recognition system combined with a deep learning model. , will not repeat them here.

Applicants have noticed that different technical fields have different requirements for data analysis. Correspondingly, the query instructions in the natural language form input by the user and the database query analysis statements obtained by the analysis are quite different, if the same semantic analysis algorithm or word is used. Semantic parsing of the library will result in slower semantic parsing and long waiting time for users.

Therefore, in order to improve the speed of the semantic analysis, in the embodiment, a plurality of semantic analysis models for different technical fields may be pre-configured in the data analysis device 100. When parsing the query instruction in the natural language form into the database query analysis statement, the data analysis device 100 may parse according to the technical domain of the data to be analyzed using different semantic analysis models. For example, for different fields such as meteorology, hydrology, finance, etc., a plurality of semantic analysis models are pre-configured in the data analysis device 100, and different semantic analysis models are configured to adopt different analysis algorithms or the thesaurus to the natural language of the corresponding technical field. Perform a quick analysis.

Specifically, in step S110, after determining the data to be analyzed, the data analysis device 100 further needs to determine a target technical field to which the data to be analyzed belongs.

In an example, the data analysis device 100 may display a plurality of selectable technical field options to the user through the mobile terminal, and respond to the user operation, and use the technical field selected by the user as the target technical field of the data to be analyzed. . For example, the data analysis device 100 displays a plurality of selection items of weather, hydrology, finance, etc. to the user through the mobile terminal after determining the data to be analyzed, and when the user selects the “financial” option, the data analysis device 100 The financial field is determined as the target technical field of the data to be analyzed.

Alternatively, in another example, the data analysis device 100 may extract a file name of the data to be analyzed and a name of each data item in the data to be analyzed, and a keyword in the file name and the data item name. Statistics are performed, and the target technical field described in the data to be analyzed is determined according to the statistical result. For example, when the file name of the data to be analyzed selected by the user is “revenue and expenditure data statistics table”, or when the name of the data item in the data to be analyzed includes “revenue”, “expenditure”, “balance”, and the like, The data analysis device 100 may determine the target technology area described by the data to be analyzed as the financial field.

Then, in step S130, the data analysis device 100 parses the natural language form query instruction according to the target technology domain determined in step S110 to obtain the database query analysis statement. .

In this way, for the data to be analyzed in different fields, the data analysis device 100 uses the corresponding semantic analysis model for parsing, which can effectively improve the parsing speed of the query instruction in the natural language form, thereby reducing the waiting time of the user in the query process. Improve the user experience.

In some embodiments, the data analysis device 100 records the corresponding database query analysis statement according to the query instruction in the natural language form input by the user, and records the query instruction and the parsed database query analysis statement to a semantic comparison. table. When other users input the query instruction in the natural language form, it first searches whether the query instruction currently input by the user has been recorded in the semantic comparison table.

If it is checked that the query instruction currently input by the user has been recorded in the semantic comparison table, indicating that the query instruction has been parsed, the data analysis device 100 directly retrieves the query instruction directly from the semantic comparison table. Corresponding database query analysis statement.

If it is detected that the query instruction currently input by the user is not recorded in the semantic comparison table, indicating that the query instruction has not been parsed, the data analysis device 100 adopts a preset semantic analysis model to query the natural language form. Analyze and get the corresponding structured database query analysis statement.

Preferably, the database analysis device 100 searches for the database query analysis statement corresponding to the query instruction in the semantic comparison table, and sequentially performs the query instruction currently input by the user and the query instruction in the semantic comparison table. Similarity analysis.

When the query instruction that is similar to the query instruction currently input by the user is greater than the first preset threshold from the semantic comparison table, the database query analysis statement corresponding to the query instruction is directly retrieved. The first preset threshold may be set to 95%.

If the similarity between the query instruction in the semantic comparison table record and the query instruction currently input by the user does not reach the first preset threshold, the query instruction in the semantic comparison table record is similar to the query instruction currently input by the user. The query instruction whose degree is greater than the second preset threshold is based on the database query analysis statement corresponding to the query instruction with the highest similarity among the query instructions having the similarity greater than the second threshold, and is adjusted according to the query instruction currently input by the user. The database query analysis statement uses the adjusted query analysis statement as a database query analysis statement corresponding to the query instruction currently input by the user. The second preset threshold may be set to 80%.

If the similarity between the query instruction in the semantic reference record and the query instruction currently input by the user does not reach the second preset threshold, the data analysis device uses the preset semantic analysis model to query the natural language form. Analyze and get the corresponding structured database query analysis statement.

In this way, the query instruction currently input by the user is first compared with the previously interpreted query instruction, so as to directly use the structured database query analysis statement that has been parsed before or on the basis of the database query analysis statement obtained before parsing. Adjustment and modification can effectively avoid repeated analysis of the same "query command" and improve the operating efficiency of the system.

In some embodiments, the semantic comparison table further records a heat value used by each query instruction, and the database query analysis statement of the record in the semantic comparison table is selected in the query instruction similarity matching process. If yes, the heat value of the query instruction corresponding to the database query analysis statement is increased. When the query instruction currently input by the user is matched and matched with the query instruction recorded in the semantic comparison table, the query instruction of the record in the semantic comparison table and the current user are sequentially sequentially according to the order of the heat value. The input query instructions are compared. In this way, the query instruction with a higher frequency of use in the semantic comparison table can preferentially match the current input query instruction of the user, which can improve the matching speed.

Step S140: Determine a data display manner according to the query instruction, and generate a visual analysis result of the statistical data according to the data display manner.

After obtaining the data to be analyzed, the processor 130 of the data analysis device 100 determines whether the specified data display mode is included in the query instruction.

When the specified data display mode is not included in the query instruction, the processor 130 generates a visual analysis result of the statistical data according to a default data display manner. For example, when the query instruction only includes “display sales in the last 10 months”, the default data display manner is used to display the data to be analyzed. The default data presentation mode is preset by the user, for example, a histogram, a line chart or a data table.

When the query instruction includes the specified data display mode, the processor 130 generates a visual analysis result of the statistical data according to the specified data display manner. For example, when the query instruction includes “displaying the income trend of each month in the last year by using a line graph”, the data to be analyzed is generated as a line graph according to the specified data display manner.

Optionally, the data display manner includes a graphic display manner. The graphic display manner includes one or more of a graphic style, a color quantity, a color type, and a graphic size. Graphic styles include, but are not limited to, line charts, histograms, dashboard charts, pie charts, radar charts, heat maps, funnel charts, or scatter plots.

Step S150, displaying the visual analysis result in the form of a dialog interaction in the data analysis interaction interface.

Referring to FIG. 4 again, after generating the visual analysis result, the processor 130 of the data analysis device 100 displays the analysis result on the data analysis interaction page in a manner of replying to the dialogue.

Based on the above design, in the human-machine dialog data solution provided by the embodiment, the input data or the query instruction of the user is received through the interactive interface of the human-machine dialogue, and the visualized query result is fed back to the user in a manner of replying to the dialogue, compared with the current In the technology, the user needs to use a special database query language to perform data query in a special database query interface. The solution provided in this embodiment provides a more friendly human-computer interaction mode, and does not require the user to have professional database query analysis knowledge. Convenient and intuitive

It is to be noted that other available variants of the embodiment further include that the third-party communication software configured by the user terminal 200 provides the data selection interface or the data analysis interaction interface, and the information or instructions sent by the user are A server of the three-party communication software is sent to the data analysis device 100. The visual analysis result generated by the data analysis device 100 is sent to the user terminal 200 via a server of the third-party communication software, and the third-party communication software configured on the user terminal 200 provides data analysis interaction. The interface displays the analysis results of the visualization. The third party communication software may be instant messaging software such as WeChat.

For example, in step S120, the data analysis device 100 provides a data analysis interaction interface through the third-party communication software configured on the user terminal, and receives the query instruction sent by the user terminal through the third-party communication software. Then, in step S150, the data analysis device 100 sends the generated visual analysis result to the user terminal, so that the user terminal displays the interaction in the form of a dialog interaction through the data analysis interaction interface provided by the third-party communication software. The results of the visual analysis are described. In such an arrangement, the third party communication software and its server should be considered to assist the data analysis device 100 in providing the data selection interface or data analysis interaction interface, and thus the above solution should also be considered as protection in the present disclosure. Within the scope.

Optionally, in this embodiment, the method may further include: the data analysis device 100 receives and displays a display mode update instruction input by the user on the data analysis interaction interface, where the display mode update instruction includes a data display manner. And then, according to the display mode update instruction, the data display manner is regenerated, and the visual analysis result of the statistical data is regenerated.

For example, after the income trend of each month of the most recent year is displayed by a line graph in the case exemplified in steps S120-S150, the user continues to input "convert to histogram", and the data analysis device 100 receives and displays the display mode update. instruction. Then, the data analysis device 100 generates the data to be analyzed into a histogram according to the re-determined data display manner based on the data to be analyzed that is queried in step S130. The newly generated visual analysis result is then displayed on the data analysis interactive interface.

Optionally, in this embodiment, the method may further include the data analysis device 100 receiving and displaying a statistical range update instruction input by the user on the data analysis interaction interface, and then re-determining according to the statistical range update instruction or Modify the statistical range corresponding to the current visual analysis result and generate a new visual analysis result.

For example, after the income trend of each month of the most recent year is displayed by the line graph in the case exemplified in steps S120-S150, the user continues to input "remove the first quarter data", and the data analysis device 100 receives and displays the statistical range. Update instructions. Then, the data analysis device 100 generates a new visual analysis result according to the re-determined statistical range on the basis of the data to be analyzed that is queried in step S130. The newly generated visual analysis result is then displayed on the data analysis interactive interface.

Optionally, in this embodiment, the method may further include: the data analysis device 100 receiving, by the user, the selection of the visual analysis result on the data analysis interaction interface, and performing the selected visual analysis result on the selected Store or send. For example, the data analysis device 100 receives a user click on the displayed line graph on the data analysis interaction interface to display at least one operational option (eg, save, send, etc.). The generated analysis result is stored or transmitted in response to the user selecting the operation option.

Optionally, in this embodiment, the method may further include: the data analysis device 100 records a query instruction input by the user on the data analysis interaction interface, and the data analysis system generates a visual analysis in response to the query instruction. As a result, an analysis interaction record is generated. The analytical interaction record is stored or transmitted in response to user operations.

Optionally, in the embodiment, the method may further include: the data analysis device 100 receiving and displaying a default data presentation mode configuration instruction input by the user on the data analysis interaction interface, where the default data display mode configuration instruction is The user-specified default data display mode is configured, and the default data display mode of the visual analysis result is configured according to the default data display mode configuration instruction. For example, the user may input “default line graph display data” in the data analysis interaction interface, and the data analysis device 100 receives and displays the default data display mode configuration instruction, and visually analyzes the default data according to the instruction. The presentation mode is configured to be displayed by a line chart.

In some embodiments, the database analysis device 100 may display an error correction feedback option on the analysis interaction interface after displaying the visual analysis result. When the user finds that the visual analysis result fed back by the database analysis device 100 is not ideal, the error correction instruction may be sent to the data analysis device 100 by selecting the error correction feedback option. After receiving the error correction instruction, the database analysis device 100 re-analyzes the user input query instruction or re-displays another query instruction whose similarity is above the first preset threshold from the semantic comparison table. For the user to select, and after the user re-analyzes the semantics or the user reselects the query instruction, a new visual analysis result is generated and displayed to the user.

Based on the above design, the method provided by the embodiment provides a friendly human-machine dialogue interaction interface. In the process of data analysis, the interaction process is intuitive and concise, so that users without professional knowledge can easily use the data. Analysis function.

If the user performs the same query action for a certain data to be analyzed in a short time, the data analysis device 100 may repeatedly perform the same query analysis action multiple times in a short time. Therefore, in this embodiment, before performing step S130, the file identifier of the data to be analyzed and the keyword in the query instruction are hashed to obtain a query hash value. Then find out if the visual analysis results associated with the query hash value have been generated. The file identifier of the data to be analyzed may be metadata of the file to be analyzed or other identifier information that identifies the unique identity of the data to be analyzed.

If the result of the visual analysis associated with the query hash value is queried, the same query statistics operation has been performed for the same data to be analyzed, and the data analysis device 100 directly jumps without performing the query statistics operation again. Go to step S150, and display the queryed visual analysis result in the form of dialog interaction in the data analysis interaction interface.

If the result of the visual analysis associated with the query hash value is not queried, the query statistical operation required by the query instruction has not been executed for the same data to be analyzed, and the data analysis device 100 performs step S130 according to the The query instruction performs statistical analysis on the data to be analyzed to obtain corresponding statistical data. After obtaining the visual analysis result according to the statistical data in step S150, the query hash value is associated with the obtained visual analysis result. In this way, the next time the same statistical operation is performed, the generated visual analysis result can be directly queried according to the query hash value.

For example, the user sends a query instruction for "representing the income trend of each month of the most recent year in a line graph" for the data file A 20 minutes ago, then the data analysis device pairs the metadata according to the data file A, the keyword " The line chart, the keyword "every month of the most recent year" and the keyword "income trend" are hashed, get the query hash value X, and then find out whether it has been associated with the query hash value X. Visualize the results of the analysis.

Since the user did not send a query instruction for displaying the income trend of each month in the last year by using a line graph A, the data analysis result corresponding to the query hash value A is not generated, and the data is not generated. The analysis device 100 performs analysis according to the query instruction input by the user, generates a visual analysis result B and presents it to the user, and associates the visual analysis result B with the query hash value.

Then, after 20 minutes, the user again sends a query instruction for "displaying the income trend of each month of the most recent year with a line graph" for the data file A, then the data analysis device pairs the metadata according to the data file A, the keyword " The line chart, the keyword "every month of the most recent year" and the keyword "income trend" are hashed, get the query hash value X, and then find out whether it has been associated with the query hash value X. Visualize the results of the analysis.

Since the visual query result B associated with the query hash value X has been previously generated, the data analysis device 100 can query the visual analysis result B according to the query hash value X, and the data analysis device 100 directly displays the visual analysis result. B feeds back the user instead of performing the statistical analysis based on the query command again.

Further, in this embodiment, in order to store the excessive visual analysis result, the data analysis device 100 may clear the storage time longer than the second preset duration every interval of the first preset duration. The visual analysis result is described, wherein the first preset duration is greater than the second preset duration. For example, the data analysis device 100 is aware of the visual analysis results stored on the data analysis device 100 for more than 20 minutes every 5 hours.

Referring to FIG. 5, the embodiment further provides a human-machine dialog data analyzing apparatus 110 applied to the data analyzing apparatus 100 shown in FIG. 1. The apparatus includes a file determining module 111, an instruction acquiring module 112, and a statistical analyzing module 113. The result generation module 114 and the result display module 115.

The file determining module 111 is configured to determine data to be analyzed.

In this embodiment, the file determining module 111 may be configured to perform step S110 shown in FIG. 2, and the description of the step S110 may be referred to with a specific description of the file determining module 111.

The instruction obtaining module 112 is configured to provide a data analysis interaction interface, and receive and display a query instruction input by the user on the data analysis interaction interface, where the query instruction includes the query content.

In this embodiment, the instruction obtaining module 112 may be configured to perform the step S120 shown in FIG. 2, and the description of the step S120 may be referred to in the specific description of the instruction acquiring module 112.

The statistical analysis module 113 is configured to perform statistical analysis on the data to be analyzed according to the query instruction, to obtain corresponding statistical data.

In this embodiment, the statistical analysis module 113 may be configured to perform the step S130 shown in FIG. 2, and the specific description of the statistical analysis module 113 may refer to the description of the step S130.

The result generating module 114 is configured to determine a data display manner according to the query instruction, and generate a visual analysis result of the statistical data according to the data display manner.

In this embodiment, the result generating module 114 may be configured to perform step S140 shown in FIG. 2, and a detailed description of the result generating module 114 may refer to the description of the step S140.

The result display module 115 is configured to display the visual analysis result on the data analysis interaction interface.

In this embodiment, the result display module 115 can be configured to perform the sub-step S150 shown in FIG. 2, and a detailed description of the result display module 115 can refer to the description of the sub-step S150.

Optionally, in this embodiment, the file determining module 111 is specifically configured to provide a data selection interface; in response to the user's operation in the data selection interface, the selected data is used as data to be analyzed.

Optionally, in the embodiment, the instruction obtaining module 112 is specifically configured to receive and display text information input by the user, and display the query instruction in a dialog interaction manner; or receive the voice input by the user through the voice input device. Information, converting the voice information into a text instruction and displaying the text instruction in the form of a dialog interaction.

Optionally, in the embodiment, the instruction obtaining module 112 is specifically configured to provide a data analysis interaction interface by using a third-party communication software configured on the user terminal, and receive the query sent by the user terminal by using the third-party communication software. instruction;

The result display module is specifically configured to send the generated visual analysis result to the user terminal, so that the user terminal displays the visual analysis result in the form of a dialog interaction through a data analysis interaction interface provided by the third-party communication software. .

In summary, the human-machine dialog data analysis method and apparatus provided by the present disclosure receive a query instruction of the user in the interface data by providing a friendly data analysis interaction interface, and perform data analysis according to the query instruction, and then use the human-machine The form of the dialog displays the analysis results on the analysis interaction interface. In this way, interaction through human-machine dialogue makes the operation of data analysis more concise and intuitive, and the interaction mode is more user-friendly, which reduces the threshold for using data analysis functions.

It is to be understood that the term "comprises", "comprising" or any other variations thereof is intended to encompass a non-exclusive inclusion, such that a process, method, article, or device that comprises a It also includes other elements that are not explicitly listed, or elements that are inherent to such a process, method, item, or device. An element that is defined by the phrase "comprising a ..." does not exclude the presence of additional equivalent elements in the process, method, item, or device that comprises the element.

The above is only the specific embodiment of the present disclosure, but the scope of the present disclosure is not limited thereto, and any person skilled in the art can easily think of changes or substitutions within the technical scope of the disclosure. It should be covered within the scope of protection of the present disclosure. Therefore, the scope of protection of the present disclosure should be determined by the scope of the claims.

Industrial applicability

The human-machine dialog data analysis method and apparatus provided by the embodiments of the present disclosure receive a query instruction of the user in the interface data by providing a friendly data analysis interaction interface, and perform data analysis according to the query instruction, in the form of human-machine dialogue. The analysis result is displayed on the analysis interaction interface. In this way, interaction through human-machine dialogue makes the operation of data analysis more concise and intuitive, and the interaction mode is more user-friendly, which reduces the threshold for using data analysis functions.

Claims (20)

A human-machine dialogue data analysis method is applied to a data analysis device, and the method includes: Determining the data to be analyzed; Providing a data analysis interaction interface, receiving a query instruction input by the user on the data analysis interaction interface, and displaying the query instruction in a dialog interaction manner, wherein the query instruction includes the query content; Performing statistical analysis on the data to be analyzed according to the query instruction, and obtaining corresponding statistical data; Determining, according to the query instruction, a data display manner, and generating a visual analysis result of the statistical data according to the data display manner; The visual analysis result is displayed in the form of a dialog interaction at the data analysis interaction interface. The method according to claim 1, wherein the step of determining data to be analyzed comprises: Providing a data selection interface; In response to the user's operation on the data selection interface, the selected data is taken as the data to be analyzed. The method according to claim 1, wherein said displaying said query instruction in the form of a dialog interaction comprises any one of the following: Receiving and displaying text information input by the user, and displaying the query instruction in the form of a dialog interaction; or The voice information input by the user through the voice input device is received, the voice information is converted into a text command, and the text command is displayed in the form of a dialog interaction. The method according to claim 1, wherein the step of providing a data analysis interaction interface, receiving a query instruction input by the user on the data analysis interaction interface, and displaying the query instruction in the form of a dialog interaction includes : Providing a data analysis interaction interface by using a third-party communication software configured on the user terminal, and receiving a query instruction sent by the user terminal through the third-party communication software; The step of displaying the visual analysis result in the form of a dialog interaction in the data analysis interaction interface includes: Sending the generated visual analysis result to the user terminal, so that the user terminal displays the visual analysis result in the form of a dialog interaction through a data analysis interaction interface provided by the third-party communication software. The method according to claim 1, wherein the query instruction comprises a statement in a natural language form; and the step of performing statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, include: Parsing the statement in the natural language form to obtain a database query analysis statement; Performing statistical analysis on the data to be analyzed according to the database query analysis statement to obtain corresponding statistical data. The method of claim 5 wherein: The step of determining data to be analyzed includes: Determining the data to be analyzed, and determining a target technical field to which the data to be analyzed belongs; The step of parsing the statement in the natural language form to obtain a database query analysis statement includes: Determining, by the corresponding language analysis model, the query instruction of the natural language form according to the target technical field to obtain the database query analysis statement. The method according to claim 6, wherein the step of determining the target technical field of the data to be analyzed comprises: Responding to user operations, using the technical field selected by the user as the target technical field; or Extracting a file name of the data to be analyzed and a name of each data item in the data to be analyzed, performing statistics on the keyword in the file name and the data item name, and determining the data to be analyzed according to the statistical result. Target technology area. The method according to claim 1, wherein the step of determining a data presentation manner according to the query instruction and generating a visual analysis result of the statistical data according to the data display manner comprises: Determining whether the specified data display mode is included in the query instruction; When the specified data display mode is not included in the query instruction, the visual analysis result of the statistical data is generated according to a default data display manner; When the specified data display mode is included in the query instruction, the visual analysis result of the statistical data is generated according to the specified data display manner. The method of claim 8 further comprising: Receiving and displaying a default data display mode configuration instruction input by the user in the data analysis interaction interface, where the default data display mode configuration instruction includes a default data display mode specified by the user, and configuring a visual analysis result according to the default data display mode configuration instruction. The default data display method. The method according to claim 8, wherein the data display mode comprises a graphic display mode; the graphic display mode comprises one or more of a graphic style, a color quantity, a color type, and a graphic size; , including: line chart, histogram, dashboard chart, pie chart, radar chart, heat map, funnel chart or scatter chart. The method of claim 1 further comprising: Receiving and displaying a display mode update instruction input by the user on the data analysis interaction interface, where the display mode update instruction includes a data display mode; And re-generating the visual analysis result of the statistical data according to the data display manner in the display mode update instruction; The newly generated visual analysis result is displayed on the data analysis interactive interface. The method of claim 1 further comprising: Receiving and displaying a statistical range update instruction input by the user on the data analysis interaction interface; The statistical range corresponding to the current visual analysis result is re-determined or modified according to the statistical range update instruction, and a new visual analysis result is generated. The method of claim 1 further comprising: Receiving, by the user, the selection of the visual analysis result on the data analysis interaction interface, and storing or transmitting the selected visual analysis result. The method of claim 1 further comprising: Recording a query instruction input by the user on the data analysis interaction interface, and the data analysis system generates a visual analysis result in response to the query instruction, and generates an analysis interaction record; The analytical interaction record is stored or transmitted in response to user operations. The method according to claim 1, wherein before the step of performing statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, the method further includes: Performing a hash calculation on the file identifier of the data to be analyzed and the keyword in the query instruction to obtain a query hash value; Find if the visual analysis result associated with the query hash value has been generated; If the visual analysis result associated with the query hash value is queried, performing the step of displaying the visual analysis result in the form of a dialog interaction in the data analysis interaction interface; If the result of the visual analysis associated with the query hash value is not queried, performing a statistical analysis on the data to be analyzed according to the query instruction to obtain corresponding statistical data, and obtaining the corresponding statistical data according to the statistical data After visualizing the analysis result, the visual analysis result is stored, and the query hash value is associated with the obtained visual analysis result. The method of claim 14, wherein the method further comprises: And displaying, by the first preset duration, the visual analysis result that the storage time exceeds the second preset duration, wherein the first preset duration is greater than the second preset duration. A device for analyzing human-machine dialogue data is applied to a data analysis system, wherein the device comprises: a file determination module configured to determine data to be analyzed; The instruction acquisition module is configured to provide a data analysis interaction interface, and receive and display a query instruction input by the user on the data analysis interaction interface, where the query instruction includes the query content; The statistical analysis module is configured to perform statistical analysis on the data to be analyzed according to the query instruction, to obtain corresponding statistical data; a result generating module, configured to determine a data display manner according to the query instruction, and generate a visual analysis result of the statistical data according to the data display manner; The result display module is configured to display the visual analysis result on the data analysis interaction interface. The device according to claim 17, wherein the instruction acquisition module is specifically configured to receive and display text information input by a user, and display the query instruction in a dialog interaction form; or receive the user input through a voice input device. The voice information converts the voice information into a text command and displays the text command in the form of a dialog interaction. The device according to claim 17, wherein the instruction acquisition module is specifically configured to provide a data analysis interaction interface through a third-party communication software configured on the user terminal, and receive the user terminal to send through the third-party communication software. Query instruction The result display module is specifically configured to send the generated visual analysis result to the user terminal, so that the user terminal displays the visual analysis result in the form of a dialog interaction through a data analysis interaction interface provided by the third-party communication software. . A readable storage medium, wherein the readable storage medium stores machine executable instructions that, when executed by a processor, cause the processor to implement any of claims 1-16 The method described in the item.

Images