CN110580259A - Customer demand mining method and equipment based on process management big data - Google Patents

Abstract

The purpose of this application is to provide a customer demand mining method and equipment based on process management big data. The computer equipment selects at least three reference users based on the basic user information of the target user, obtains the historical service items of each of the reference users and determines A plurality of first preselected service items, determining the historical service items of the target user based on the user basic information and determining a plurality of second preselected service items, and then determining information about the target user based on the first preselected service items and the second preselected service items Customized service items for the user's customer needs. This application can realize the precise mining of customer needs, and improve the conversion rate of recommended traffic of intellectual property consulting services and the click-through rate of recommended services.

Description

Customer demand mining method and equipment based on process management big data

technical field

This application relates to the computer field, and in particular to a customer demand mining technology based on process management big data.

Background technique

With the development of information technology, business users rely more and more on consulting services of professional institutions (such as intellectual property services) to improve work efficiency. However, with the continuous enrichment of consulting service items, it becomes more and more difficult to provide users with timely and accurate services. In addition, in some cases, users are not very clear about their specific needs and whether their own conditions are suitable for certain projects. At this time, providing quick feedback will greatly improve the user's consultation efficiency and improve user satisfaction. .

Contents of the invention

One purpose of this application is to provide a customer demand mining method based on process management big data.

According to one aspect of the present application, a method for mining customer needs based on process management big data is provided, the method includes the following steps:

a. Based on the basic user information of the target user, selecting at least three reference users with the highest matching degree between the basic user information and the basic user information of the target user from the user database;

b. Acquiring the historical service items of each of the reference users, and applying the historical service items of the reference users to the first customization model to determine a plurality of first pre-selected service items;

c determining the historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model to determine a plurality of second pre-selected service items; and

d Applying the plurality of first pre-selected service items, the plurality of second pre-selected service items and the user basic information of the target user to a third customization model to determine customized services related to the customer needs of the target user project.

In some embodiments, the step b includes:

Obtain historical service items of each of the reference users, and apply the historical service item information of the reference users to the first customized model; and

Based on the output of the first customization model, a plurality of first preselected service items are determined, wherein the intersection of the plurality of first preselected service items and the historical service of the target user is empty.

In some embodiments, the step c includes:

determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model; and

A plurality of second preselected service items are determined based on the output of the second customization model, wherein the intersection of the plurality of second preselected service items and the historical service of the target user is empty.

In some embodiments, the step b includes:

Obtaining historical service items of each of the reference users, and applying the historical service items and user attribute information of the reference users to the first customization model to determine a plurality of first pre-selected service items;

Said step c comprises:

determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items and user attribute information of the target user to a second customization model to determine a plurality of second preselected service items; and

Said step d comprises:

applying the plurality of first pre-selected service items, the plurality of second pre-selected service items, the user basic information of the target user, and the user attribute information of the target user to a third customized model to determine information about the Customized service items for the customer needs of the target users.

In some embodiments, the step c includes:

determining historical service items of the target user based on the user basic information of the target user, and applying the historical service item information of the target user to a second customized model; and

A plurality of second preselected service items are determined based on the existing intellectual property information of the target user.

In some embodiments, the step d includes:

applying the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the user basic information of the target user to a third customized model; and

Based on the existing intellectual property information of the target user, a customized service item related to the customer demand of the target user is determined.

In some embodiments, step a includes:

a1 obtain the user operation data of the target user in the IVR system;

a2 determining the desired service items of the target user based on the user operation data; and

a3 based on the basic user information of the target user, selecting at least three reference users whose basic user information matches the basic user information of the target user from the user database; and

Said step d comprises:

applying the plurality of first preselected service items, the plurality of second preselected service items, the desired service items of the target user, and the user basic information of the target user to a third customized model to determine the Customized service items for the customer needs of the target users.

In some embodiments, the step a2 includes:

Applying the user operation data to a fourth customization model to determine the desired service items of the target user.

In some embodiments, the third customized model is generated based on a residual feed-forward convolutional network.

According to another aspect of the present application, a computer device is provided, and the computer device includes:

processor; and

A memory arranged to store computer-executable instructions that, when executed, cause the processor to:

Based on the basic user information of the target user, selecting at least three reference users with the highest matching degree between the basic user information and the basic user information of the target user from the user database;

Obtaining the historical service items of each of the reference users, and applying the historical service items of the reference users to the first customization model to determine a plurality of first pre-selected service items;

determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model to determine a plurality of second preselected service items; and

applying the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the target user's basic user information to a third customization model to determine customized service items related to the target user's customer needs .

Compared with the existing technology, this application is based on the accumulated data of customer process management, and uses feedback learning algorithm to structure the text information such as consulting reports and logs to obtain more comprehensive and accurate customer information. Service and other two aspects to mine recommendable services, and specifically introduce a residual feed-forward convolutional network to correct errors in recommended services (such as intellectual property services), thereby further improving the accuracy of recommended services. This application can realize the precise mining of customer needs, and improve the conversion rate of recommended traffic of intellectual property consulting services and the click-through rate of recommended services.

Description of drawings

Other characteristics, objects and advantages of the present application will become more apparent by reading the detailed description of non-limiting embodiments made with reference to the following drawings:

Fig. 1 shows the process flow of a kind of customer demand mining method based on process management big data according to one embodiment of the present application;

Fig. 2 shows the input-output relationship between the first customized model, the second customized model and the third customized model according to another embodiment of the present application;

Fig. 3 shows the structure of the neural network of matching similar enterprises in one embodiment of the present application;

Fig. 4 shows the structure of the neural network of matching relevant business of one embodiment of the present application;

Fig. 5 shows the structure of the neural network in the 3rd customization model of an embodiment of the present application;

Fig. 6 shows the computer equipment of an embodiment of the present application;

The same or similar reference numerals in the drawings represent the same or similar components.

Detailed ways

The application will be described in further detail below in conjunction with the accompanying drawings.

In a typical configuration of the present application, the terminal, the device serving the network and the trusted party all include one or more processors (CPUs), input/output interfaces, network interfaces and memory.

Memory may include non-permanent storage in computer readable media, in the form of random access memory (RAM) and/or nonvolatile memory such as read only memory (ROM) or flash RAM. Memory is an example of computer readable media.

Computer-readable media, including both permanent and non-permanent, removable and non-removable media, can be implemented by any method or technology for storage of information. Information may be computer readable instructions, data structures, modules of a program, or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read only memory (ROM), Electrically Erasable Programmable Read-Only Memory (EEPROM), Flash memory or other memory technology, Compact Disc Read-Only Memory (CD-ROM), Digital Versatile Disc (DVD) or other optical storage, Magnetic tape cartridge, tape disk storage or other magnetic storage device or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

The equipment referred to in this application includes, but is not limited to, user equipment, network equipment, or equipment formed by integrating user equipment and network equipment through a network. The user equipment includes but is not limited to any mobile electronic product that can perform human-computer interaction (such as human-computer interaction through a touch panel) with the user, such as a smart phone, a tablet computer, etc., and the mobile electronic product can use any operation System, such as android operating system, iOS operating system, etc. Wherein, the network device includes an electronic device that can automatically perform numerical calculation and information processing according to preset or stored instructions, and its hardware includes but is not limited to microprocessors, application-specific integrated circuits (ASICs), programmable logic device (PLD), field programmable gate array (FPGA), digital signal processor (DSP), embedded device, etc. The network equipment includes but is not limited to a computer, a network host, a single network server, a plurality of network server sets or a cloud composed of multiple servers; here, the cloud is composed of a large number of computers or network servers based on Cloud Computing, Among them, cloud computing is a kind of distributed computing, a virtual supercomputer composed of a group of loosely coupled computer sets. The network includes, but is not limited to, the Internet, a wide area network, a metropolitan area network, a local area network, a VPN network, a wireless ad hoc network (Ad Hoc network) and the like. Preferably, the device may also be a program running on the user device, network device, or a device formed by integrating user device and network device, network device, touch terminal or network device and touch terminal through a network.

Of course, those skilled in the art should understand that the above-mentioned equipment is only an example, and other existing or future equipment that may be applicable to this application should also be included in the scope of protection of this application, and it is included in this application by reference. this.

In the description of the present application, "plurality" means two or more, unless otherwise specifically defined.

According to one aspect of the present application, a method for mining customer needs based on process management big data is provided, the method includes step S100, step S200, step S300 and step S400. Taking a computer device as an example, the customer demand mining method based on process management big data will be described in detail below.

Referring to FIG. 1 , in step S100 , the computer device selects at least three reference users whose basic user information matches the basic user information of the target user from the user database based on the basic user information of the target user. Among them, the target user is the object of this customer demand mining based on process management big data, generally an enterprise. The basic user information of the target user can be collected by business personnel from the enterprise, or obtained from the business information query website; in some embodiments, the basic user information of the target user includes the name of the company, type of business, registered capital, date of establishment, residence, business scope etc., but the application is not limited thereto. The user database contains the user basic information of multiple users, and also includes the historical service project data of these multiple users, such as the consulting activities that each user has conducted or the status of related consulting projects (such as but not limited to, in progress, project success, project failure, etc.). Matching with reference users can be done based on a neural network. For example, referring to FIG. 3 , the data D used in the matching process mainly includes the basic information of the enterprise and the character portrait of the enterprise by the business personnel. During the training of the model, all the data is randomly divided into two parts, 70% for training and 30% for testing. In the training data, two company data are randomly selected and input into the neural network, and the neural network is trained using the scores of business people. During detection, the data D _Target of the target company and the data D _Reference ([D _Target ; D _Reference ]) of any other user in the database are input into the neural network to obtain a similarity score S _client . The neural network is a four-layer fully connected network, uses a cross-entropy function as a loss function, and uses a gradient descent method for training. After traversing all users, three reference users ClientR ₁ , ClientR ₂ , and ClientR ₃ that are closest to the target user in terms of user attributes (the largest S _client ) can be selected.

In step S200, the computer device acquires the historical service items of each reference user, and applies the historical service items of the reference user to the first customized model to determine a plurality of first pre-selected service items. Wherein, the first customized model is used for voting and evaluating the historical service items of the reference user, so as to recommend consulting service items suitable for the target user. The first customized model is a known data model, such as neural network, decision tree, linear regression and other data models. Preferably, in step S200, the computing device first acquires the historical service items of each of the reference users, and applies the historical service item information of the reference users to the first customized model; then, the computing device based on the first Customize the output of the model to determine a plurality of first preselected service items, wherein the intersection of the plurality of first preselected service items and the historical services of the target user is empty, in other words, select the consulting service items that the target user has not yet involved As the first pre-selected service item.

For example, following the previous example, use these three companies to refer to the user's historical business to select a number of first pre-selected service items. The first pre-selected service items in this example are five intellectual property services service ₁ , service ₂ , service ₃ , service ₄ , service ₅ .

Wherein, in some embodiments, the computer device acquires the historical service items of each reference user in step S200, and applies the historical service items and user attribute information of the reference user to the first customized model to determine multiple a first pre-selected service item; wherein, the user attribute information is the classification of the user, the label corresponding to the user, etc., for example, the personality portrait of the business person to the enterprise.

In step S300, the computer device determines the historical service items of the target user based on the user basic information of the target user, and applies the historical service items of the target user to the second customized model to determine a plurality of second preselected service items. The data M used in this step mainly includes historical project information of the enterprise; for intellectual property service items, the data M used may also include the existing intellectual property information of the enterprise (such as the number of patents), etc. Referring to FIG. 4, in some embodiments, the second custom model also includes a neural network. During the training of this neural network, all the data is randomly divided into two parts, 70% for training and 30% for testing. In the training data, two company data are randomly selected and input into the neural network, and the neural network is trained using the scores of business people. During detection, input the data M _Target of the target user and the data M _Reference ([M _Target ; M _Reference ]) of any other user in the database into the neural network to obtain a similarity score S _service . The neural network is also a four-layer fully connected network, and cross-entropy is selected as the loss function again, and it is trained by the method of gradient descent. After traversing all users, three reference users ClientR ₄ , ClientR ₅ , and ClientR ₆ that are closest to the target user in historical business (the largest S _service ) can be selected; the second custom model then utilizes the history of these three customers The business votes to select a plurality of second pre-selected service items. Wherein, preferably, in step S300, the computer device first determines the historical service items of the target user based on the basic user information of the target user, and applies the historical service items of the target user to the second customized model; then , the computer device determines a plurality of second pre-selected service items based on the output of the second customization model, wherein the intersection of the plurality of second pre-selected service items and the historical service of the target user is empty, in other words, the selected target The consulting service items that the user has not been involved in are used as the second pre-selected service items.

For example, in the above example, the multiple second pre-selected service items are five types of intellectual property services service ₆ , service ₇ , service ₈ , service ₉ , and service ₁₀ that the target customer has not yet involved.

Wherein, in some embodiments, the computer device determines the historical service items of the target user based on the user basic information of the target user in step S300, and applies the historical service items and user attribute information of the target user to the first 2. Customize the model to determine a plurality of second pre-selected service items; wherein, the user attribute information is the classification of the user, the label corresponding to the user, etc., such as the personality portrait of the business person for the enterprise.

In step S400, referring to FIG. 2 , the computer device applies the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the user basic information of the target user to a third customized model to determine information about Customized service items for the customer needs of the target user. For example, following the above example, multiple first preselected service items (for example, service ₁ , service ₂ , service ₃ , service ₄ , service ₅ ) output by the first customized model and multiple second preselected service items output by the second customized model Service items (such as service ₆ , service ₇ , service ₈ , service ₉ , service ₁₀ ) and user basic information of the target user are input into the third customization model to determine customized service items related to the target user's customer needs. In some embodiments, the third customized model is also a neural network model.

Preferably, in some embodiments, the third customized model is generated based on a residual feed-forward convolutional network, and the data W used includes basic information of the target enterprise, business personnel's personality portrait of the enterprise, enterprise historical project information and / or existing intellectual property information, etc., and the customized service items excavated by the above two demand mining algorithms. Continuing from the above example, the output of the third customized model is 6 recommendations (RS ₁ -RS ₆ ) that are most likely to be selected. In a specific embodiment, the neural network in the third customized model is a 7-layer fully connected network. Referring to Figure 5, layers 1, 2, and 3 are fully connected layers, layer 4 is a 3x1 convolutional layer, and layers 5, 6, and 7 are fully connected layers. The addition of the convolutional layer allows the network to consider not only the accuracy of a single recommendation, but more importantly, the joint consideration of the six recommended services. Compared with other data models, the combination of the first custom model and the second custom model can obtain Better recommendation accuracy. For consulting service projects, it not only involves the status of the company itself (such as the basic information of the company, the number of intellectual property rights, etc.), but also needs to consider the consulting situation of similar companies (such as industry dynamics, etc., which can reflect industry development trends, relevant departmental policies, etc. ), so the convolutional layer can combine the business conditions of companies similar to the target company and the business status of the target company itself, so as to obtain comprehensive, accurate and in-demand recommendation results. Taking the recommendation of 6 services from 10 services as an example, in some embodiments, the loss function of the above neural network is determined based on two factors of similar enterprises and the enterprise itself, expressed as the following formula:

In the above formula, "max|| _RSi ||" represents the error of the service with the largest error among the recommended services, and "min||RSi||" represents the error of the service with the smallest error among the recommended services, thus integrating the information about There are two factors that are similar to the business of the enterprise and the situation of the enterprise itself. Here, "error" refers to the error between the service predicted by the system and whether the customer actually chooses this service in the end. For example, when the system predicts that a customer has a 0.8 possibility of choosing a certain service, but the customer does not choose the service in the end, then the possibility of the user finally choosing the service is 0, and the corresponding error is 0.8; on the other hand, if the system predicts that the The customer has a 0.8 possibility of choosing the service, and in the end the customer chooses the service, then the corresponding error is 0.2.

The neural network is trained using the gradient descent algorithm. The above loss function can help the network to converge quickly, and the detection accuracy is also greatly improved.

The third customized module can be trained with real data after the system goes online, so as to obtain training data that is as real and effective as possible. Continuing from the above example, if the 10 businesses recommended by the first customization model and the second customization model are finally selected by the target customer, the feedback is considered to be 0, otherwise it is 1. After continuous long-term dynamic training, the third customization module will have the ability to correct errors for the above-mentioned demand mining algorithm modules based on similar customers and demand mining algorithm modules based on similar services, making the business recommended by the system more accurate and the customized service items provided It is more in line with the actual needs of customers.

Wherein, in some embodiments, in step S400, the computer device combines the plurality of first pre-selected service items, the plurality of second pre-selected service items, the user basic information of the target user, and the user information of the target user The attribute information is applied to the third customization model to determine the customized service items related to the customer needs of the target user; wherein, the user attribute information is the classification of the user, the label corresponding to the user, etc., such as the personality portrait of the business person for the enterprise.

Sometimes the service items expected by the user cannot be completed due to the user's own insufficient conditions. For example, in the case of intellectual property services, it is possible that an intellectual property project cannot be completed because the user has insufficient quantity of one or some types of intellectual property. In order to avoid delaying the customer's time or affecting customer experience due to such problems, in step S300, the computer device first determines the historical service items of the target user based on the basic user information of the target user, and sends the target user's The historical service item information is applied to the second customization model; then, the computer device determines a plurality of second preselected service items based on the existing intellectual property information of the target user. For example, according to the status of the user's existing intellectual property information, services that the user cannot currently handle can be excluded. On this basis, in some embodiments, the computer device first applies the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the user basic information of the target user to the first step S400. 3. Customization model; then, the computer device determines the customized service items related to the customer needs of the target user based on the existing intellectual property information of the target user.

In some cases, the user calls the consulting service institution, and interacts with an IVR (Interactive Voice Response, Interactive Voice Response) system provided by the institution to obtain part of the required information by himself, such as recommended service items. Since the IVR system can complete part of the knowledge preparation work before manual intervention and formally start providing consulting services to users, the consulting process is more purposeful, so that the IVR system can improve the consulting efficiency of users and effectively improve the user's recommendation during consulting. The conversion rate of traffic and the click-through rate of recommended services. In some embodiments, step S100 includes sub-step S101, sub-step S102 and sub-step S103. In sub-step S101, the computer device acquires user operation data of the target user in the IVR system. In sub-step S102, the computer device determines the expected service items of the target user based on the user operation data, wherein the expected service items may reflect the services that the user may most need at present. For example, in some embodiments, the computer device records the user's operations in the IVR system (for example, the user presses several number keys successively according to the voice prompt during the IVR interaction process, and the system records the user's key sequence), and according to the user's The key sequence is used to look up the table to determine the service required by the user. In some other embodiments, the system inputs the user's operations in the IVR system into another data model (may be referred to as a fourth customized model), and the data model outputs the services currently required by the user, wherein the data model is based on a decision tree, Implementations such as logistic regression, support vector machine, neural network, etc., but this application is not limited thereto. Next, in sub-step S103, the computer device selects at least three reference users whose basic user information matches the basic user information of the target user from the user database based on the basic user information of the target user. Correspondingly, in step S400, the computing device applies the plurality of first pre-selected service items, the plurality of second pre-selected service items, the desired service items of the target user, and the basic user information of the target user to the The third customization model is to determine customized service items related to the customer needs of the target user.

According to another aspect of the present application, a computer device is provided, and the computer device includes:

processor; and

A memory arranged to store computer-executable instructions that, when executed, cause the processor to:

Based on the basic user information of the target user, selecting at least three reference users with the highest matching degree between the basic user information and the basic user information of the target user from the user database;

Obtaining the historical service items of each of the reference users, and applying the historical service items of the reference users to the first customization model to determine a plurality of first pre-selected service items;

determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model to determine a plurality of second preselected service items; and

applying the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the target user's basic user information to a third customization model to determine customized service items related to the target user's customer needs .

A computer device 600 according to an embodiment of the present application is described below with reference to FIG. 6 . Wherein, the device 600 shown in FIG. 6 is only an example, and should not impose any limitation on the functions and application scope of this embodiment of the present application.

As shown in FIG. 6, device 600 takes the form of a general-purpose computing device. Components of the device 600 may include, but are not limited to: at least one processor 610, at least one storage unit 620, a bus 630 connecting different system components (including the storage unit 620 and the processor 610), a display unit 640, and the like.

Wherein, the storage unit stores program codes, and the program codes can be executed by the processor 610, so that the processor 610 executes the various exemplary methods according to the present application described in the above-mentioned part of the electronic prescription circulation processing method in this specification. Implementation steps. For example, the processor 610 may execute the steps in the above method.

The storage unit 620 may include a readable medium in the form of a volatile storage unit, such as a random access storage unit (RAM) 6201 and/or a cache storage unit 6202 , and may further include a read-only storage unit (ROM) 6203 .

The storage unit 620 may also include a program/utility 6204 having a set (at least one) of program modules 6205, such program modules 6205 including but not limited to: an operating system, one or more application programs, other program modules, and program data, each or some combination of these examples may include the implementation of the network environment.

Bus 630 may represent one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, an accelerated graphics port, a processor, or a local area using any of a variety of bus structures. bus.

Device 600 may also communicate with one or more external devices 700 (e.g., keyboards, pointing devices, Bluetooth devices, etc.), communicate with one or more devices that enable tenants to interact with the electronic device 600, and/or communicate with devices that enable tenants to interact with the electronic device 600. The electronic device 600 is capable of communicating with any device (eg, router, modem, etc.) that communicates with one or more other computing devices. Such communication may occur through input/output (I/O) interface 650 . Moreover, the electronic device 600 can also communicate with one or more networks (such as a local area network (LAN), a wide area network (WAN) and/or a public network such as the Internet) through the network adapter 660 . The network adapter 660 can communicate with other modules of the electronic device 600 through the bus 630 . It should be appreciated that although not shown, other hardware and/or software modules may be used in conjunction with electronic device 600, including but not limited to: microcode, device drivers, redundant processors, external disk drive arrays, RAID systems, tape drives And data backup storage system, etc.

It should be noted that the present application can be implemented in software and/or a combination of software and hardware, for example, it can be implemented by using an application specific integrated circuit (ASIC), a general-purpose computer or any other similar hardware devices. In one embodiment, the software program of the present application can be executed by a processor to realize the steps or functions described above. Likewise, the software program (including associated data structures) of the present application can be stored in a computer-readable recording medium such as RAM memory, magnetic or optical drive or floppy disk and the like. In addition, some steps or functions of the present application may be implemented by hardware, for example, as a circuit that cooperates with a processor to execute each step or function.

In addition, a part of the present application can be applied as a computer program product, such as a computer program instruction. When it is executed by a computer, the method and/or technical solution according to the present application can be invoked or provided through the operation of the computer. Those skilled in the art should understand that computer program instructions exist in computer-readable media in forms including but not limited to source files, executable files, installation package files, etc. Limited to: the computer directly executes the instruction, or the computer compiles the instruction and then executes the corresponding compiled program, or the computer reads and executes the instruction, or the computer reads and installs the instruction and then executes the corresponding post-installation program program. Here, a computer readable medium may be any available computer readable storage medium or communication medium that can be accessed by a computer.

Communication media includes the media whereby communication signals embodying, for example, computer readable instructions, data structures, program modules or other data are transmitted from one system to another. Communication media can include guided transmission media such as cables and wires (e.g., fiber optics, coaxial, etc.) and wireless (unguided transmission) media capable of propagating waves of energy, such as acoustic, electromagnetic, RF, microwave, and infrared . Computer readable instructions, data structures, program modules or other data may be embodied, for example, as a modulated data signal in a wireless medium such as a carrier wave or similar mechanism such as embodied as part of spread spectrum technology. The term "modulated data signal" means a signal that has one or more of its characteristics changed or set in such a manner as to encode information in the signal. Modulation can be analog, digital or mixed modulation techniques.

By way of example, and not limitation, computer-readable storage media may include volatile and nonvolatile, volatile, volatile, or Removable and non-removable media. For example, computer-readable storage media include, but are not limited to, volatile memories such as random access memories (RAM, DRAM, SRAM); and nonvolatile memories such as flash memory, various read-only memories (ROM, PROM, EPROM) , EEPROM), magnetic and ferromagnetic/ferroelectric memory (MRAM, FeRAM); and magnetic and optical storage devices (hard disks, tapes, CDs, DVDs); or other media known now or developed in the future capable of storing data for computer systems Computer readable information/data used.

Here, an embodiment according to the present application includes an apparatus comprising a memory for storing computer program instructions and a processor for executing the program instructions, wherein when the computer program instructions are executed by the processor, triggering The operation of the device is based on the foregoing methods and/or technical solutions according to multiple embodiments of the present application.

It will be apparent to those skilled in the art that the present application is not limited to the details of the exemplary embodiments described above, but that the present application can be implemented in other specific forms without departing from the spirit or essential characteristics of the present application. Therefore, the embodiments should be regarded as exemplary and not restrictive in all points of view, and the scope of the application is defined by the appended claims rather than the foregoing description, and it is intended that the scope of the present application be defined by the appended claims rather than by the foregoing description. All changes within the meaning and range of equivalents of the elements are embraced in this application. Any reference sign in a claim should not be construed as limiting the claim concerned. In addition, it is obvious that the word "comprising" does not exclude other elements or steps, and the singular does not exclude the plural. A plurality of units or means stated in the device claims may also be realized by one unit or device through software or hardware. The words first, second, etc. are used to denote names without implying any particular order.

Claims (10)

1. A customer demand mining method based on flow management big data comprises the following steps:

a, based on the user basic information of a target user, selecting at least three reference users with highest matching degree of the user basic information and the user basic information of the target user from a user database;

b, acquiring the historical service items of each reference user, and applying the historical service items of the reference users to a first customized model to determine a plurality of first pre-selected service items;

c, determining the historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model to determine a plurality of second preselected service items; and

d applying the plurality of first preselected service items, the plurality of second preselected service items, and the user base information of the target user to a third customization model to determine customized service items regarding the customer needs of the target user.

2. The method of claim 1, wherein the step b comprises:

Acquiring the historical service items of each reference user, and applying the historical service item information of the reference user to a first customized model; and

Determining a first plurality of pre-selected service items based on an output of the first customized model, wherein an intersection of the first plurality of pre-selected service items with a historical service of the target user is empty.

3. The method according to claim 1 or 2, wherein said step c comprises:

Determining the historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model; and

Determining a plurality of second pre-selected service items based on an output of the second customization model, wherein an intersection of the plurality of second pre-selected service items with a historical service of the target user is empty.

4. The method according to any one of claims 1 to 3, wherein step b comprises:

acquiring the historical service items of each reference user, and applying the historical service items and the user attribute information of the reference users to a first customized model to determine a plurality of first preselected service items;

The step c comprises the following steps:

Determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items and user attribute information of the target user to a second customization model to determine a plurality of second preselected service items; and is

The step d comprises the following steps:

applying the first plurality of preselected service items, the second plurality of preselected service items, the user base information of the target user, and the user attribute information of the target user to a third customization model to determine customized service items regarding the customer needs of the target user.

5. the method of claim 1, wherein the step c comprises:

Determining the historical service items of the target user based on the user basic information of the target user, and applying the historical service item information of the target user to a second customization model; and

Determining a plurality of second pre-selected service items based on the existing intellectual property information of the target user.

6. The method of claim 5, wherein the step d comprises:

Applying the plurality of first pre-selected service items, the plurality of second pre-selected service items, and the user base information of the target user to a third customization model; and

Determining a customized service item regarding the target user's customer needs based on the target user's existing intellectual property information.

7. the method of claim 1, wherein the step a comprises:

a1, acquiring user operation data of a target user in the IVR system;

a2 determining the expected service item of the target user based on the user operation data; and

a3, based on the user basic information of the target user, selecting at least three reference users from a user database, wherein the matching degree of the user basic information and the user basic information of the target user is the highest; and is

The step d comprises the following steps:

applying the first plurality of pre-selected service items, the second plurality of pre-selected service items, the desired service items of the target user, and the user base information of the target user to a third customization model to determine customized service items regarding the customer needs of the target user.

8. the method of claim 7, wherein the step a2 includes:

applying the user operation data to a fourth customization model to determine a desired service item for the target user.

9. the method of claim 1, wherein the third customized model is generated based on a residual feed-forward convolutional network.

10. A computer device, the computer device comprising:

A processor; and

A memory arranged to store computer executable instructions that, when executed, cause the processor to:

selecting at least three reference users with highest matching degree of the user basic information and the user basic information of the target user from a user database based on the user basic information of the target user;

acquiring the historical service items of each reference user, and applying the historical service items of the reference users to a first customized model to determine a plurality of first pre-selected service items;

Determining historical service items of the target user based on the user basic information of the target user, and applying the historical service items of the target user to a second customization model to determine a plurality of second preselected service items; and

applying the first plurality of preselected service items, the second plurality of preselected service items, and the user base information of the target user to a third customization model to determine customized service items for the customer needs of the target user.