WO2022105137A1

WO2022105137A1 - Case handling method and apparatus, and computer device and computer-readable storage medium

Info

Publication number: WO2022105137A1
Application number: PCT/CN2021/091705
Authority: WO
Inventors: 黑晓群
Original assignee: 平安普惠企业管理有限公司
Priority date: 2020-11-19
Filing date: 2021-04-30
Publication date: 2022-05-27
Also published as: CN112396325B; CN112396325A

Abstract

A case handling method and apparatus, and a computer device and a computer-readable storage medium. The method comprises: respectively determining, by using case sample data, target attributes and target attribute value division intervals for cases and case handling personnel; for each group of first case sample data in which target attribute values of corresponding case handling personnel are in the same target attribute value division interval, performing training to obtain a corresponding case success rate model; for each group of second case sample data in which target attribute values of corresponding cases are in the same target attribute value division interval, performing training to obtain a case handling personnel success rate model corresponding to this group of second case sample data; and combining the models, and recommending cases or case handling personnel by using a combined model, so as to handle cases according to a recommendation result. By means of the method, case handling personnel can be sufficiently matched with cases, thereby improving the case handling quality and the case handling efficiency.

Description

Case handling method, apparatus, computer equipment and computer-readable storage medium

This application is required to be submitted to the China Patent Office on November 19, 2020, and the application number is CN 202011306736.3, the priority of the Chinese patent application entitled "Case Processing Method, Apparatus, Medium and Electronic Device", the entire contents of which are incorporated in this application by reference.

technical field

The present application relates to the field of machine learning technology, and in particular, to a case processing method, apparatus, computer device and computer-readable storage medium.

Background technique

In the process of case handling, the assignment of cases and personnel is a very important part. At present, the assignment of cases and personnel in various fields is assigned by the supervisor according to personal judgment, and the processing personnel get the assigned tasks for processing.

technical problem

The inventor realizes that in the current process of case and personnel allocation, it is completely dependent on the personal experience of the supervisor. The supervisor makes judgments based on the volume of cases or his knowledge of the cases and personnel, and then assigns the cases to the case handlers he thinks are suitable. The case handler can only passively accept the case. Therefore, the current allocation of cases and personnel is entirely dependent on individual subjective judgments, and cases cannot be allocated reasonably and objectively, resulting in the inability of case handlers to match the case well. cases, resulting in poor case handling quality and low handling efficiency.

technical solutions

In the field of machine learning technology, in order to solve the above technical problems, the purpose of this application is to provide a case processing method, apparatus, computer equipment and computer-readable storage medium.

In the first aspect, a case handling method is provided, including:

Use the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler respectively, wherein the case sample data includes the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handler, The attribute value corresponding to the attribute of the case handler and the case handling result;

For the first case sample data in which the target attribute value of each group of corresponding case handlers is in the same target attribute value division interval, the target attribute value of the case and the target attribute value corresponding to the target attribute of the case in the first case sample data of the group are used. and the case processing results to train the case success rate model to obtain the case success rate model corresponding to the sample data of the first case in the group;

For the second case sample data in which the target attribute value of each group of corresponding cases is in the same target attribute value division interval, the target attribute of the case handler and the target attribute corresponding to the target attribute of the case handler in the second case sample data of the group are used. The attribute value and the case processing result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

Each of the case success rate models and each of the case handler success rate models are combined, and the combined model is used to recommend a case or a case handler, so as to handle the case according to the recommendation result.

In a second aspect, a case processing device is provided, including:

The determining module is configured to use the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler respectively, wherein the case sample data includes the attribute of the case, the attribute value corresponding to the attribute of the case, Attributes of case handlers, attribute values corresponding to the attributes of case handlers, and case handling results;

The first training module is configured to use the target attribute of the case, the case in the group of the first case sample data, and the The target attribute value corresponding to the target attribute and the case processing result are used to train the case success rate model, and the case success rate model corresponding to the first case sample data of the group is obtained;

The second training module is configured to use the target attributes of the case handlers, the The target attribute value corresponding to the target attribute of the handler and the case handling result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

The combination and recommendation module is configured to combine each of the case success rate models with each of the case handler success rate models, and use the combined model to recommend cases or case handlers, so as to process cases according to the recommendation results.

In a third aspect, there is provided a computer device including a memory and a processor, the memory for storing a case processing program of the processor, the processor being configured to execute the following via executing the case processing program Processing: Use the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler respectively, wherein the case sample data includes the case attribute, the attribute value corresponding to the case attribute, and the case handler's attribute value. Attributes, the attribute values corresponding to the attributes of the case handlers, and the case handling results; for the first case sample data whose target attribute values of each group of corresponding case handlers are in the same target attribute value division, the first case sample of the group is used. The target attribute of the case in the data, the target attribute value corresponding to the target attribute of the case, and the case processing result are used to train the case success rate model to obtain the case success rate model corresponding to the first case sample data in the group; for each group of corresponding cases The second case sample data whose target attribute value is in the same target attribute value division interval is carried out by using the target attribute of the case handler, the target attribute value corresponding to the target attribute of the case handler and the case processing result in the second case sample data of the group. The training of the success rate model of case handlers is to obtain the case handler success rate model corresponding to the second case sample data in the group; the success rate models for each case and the success rate models for each case handler are combined, and after the combination is used The model makes recommendations for cases or case handlers so that cases can be handled according to the recommendation results.

In a fourth aspect, a computer-readable storage medium storing computer-readable instructions is provided, and a case-handling program is stored thereon, and when the case-handling program is executed by a processor, the following processing is realized: using cases in a sample set The sample data respectively determines target attributes and target attribute values for the case and the case handler, wherein the case sample data includes the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handlers, and the attributes of the case handlers Corresponding attribute values and case processing results; for the first case sample data in which the target attribute values of each group of corresponding case handlers are in the same target attribute value division, the target attributes, The target attribute value corresponding to the target attribute of the case and the case processing result are used to train the case success rate model to obtain the case success rate model corresponding to the first case sample data in this group; the target attribute value for each group of corresponding cases is in the same target The second case sample data in which the attribute value is divided into intervals, use the target attribute of the case handler in the second case sample data of the group, the target attribute value corresponding to the target attribute of the case handler, and the case processing result to carry out the success rate model of the case handler. training to obtain the case handler success rate model corresponding to the second case sample data in the group; combine each of the case success rate models with the case handler success rate models, and use the combined model to conduct case or case handler success rates. , so that the case can be handled according to the recommendation results.

beneficial effect

The above case handling method, device, computer equipment and computer-readable storage medium, on the one hand, by constructing a case success rate model, it is possible to automatically recommend a case handler suitable for handling the case to be handled; By building a case handler success rate model, the case handler can automatically recommend pending cases suitable for the case handler to handle. Therefore, the case handler can be well matched with the case, thereby improving case handling. Quality and case handling efficiency.

It is to be understood that the foregoing general description and the following detailed description are exemplary only and do not limit the application.

Description of drawings

FIG. 1 is a schematic diagram of a system architecture of a case processing method according to an exemplary embodiment.

Fig. 2 is a flow chart of a case processing method according to an exemplary embodiment.

FIG. 3 is a flowchart of steps before step 230 according to an embodiment shown in the embodiment of FIG. 2 .

Fig. 4 shows a schematic flow chart of the basic flow of the case processing method applied to the collection field according to an exemplary embodiment.

Fig. 5 is a block diagram of a case processing apparatus according to an exemplary embodiment.

Fig. 6 is an exemplary block diagram of a computer device implementing the above case processing method according to an exemplary embodiment.

FIG. 7 is a computer-readable storage medium for implementing the above case processing method according to an exemplary embodiment.

Embodiments of the present invention

Exemplary embodiments will be described in detail herein, examples of which are illustrated in the accompanying drawings. Where the following description refers to the drawings, the same numerals in different drawings refer to the same or similar elements unless otherwise indicated. The implementations described in the illustrative examples below are not intended to represent all implementations consistent with this application. Rather, they are merely examples of apparatus and methods consistent with some aspects of the present application as recited in the appended claims.

Furthermore, the drawings are merely schematic illustrations of the present application and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and thus their repeated descriptions will be omitted. Some of the block diagrams shown in the figures are functional entities that do not necessarily necessarily correspond to physically or logically separate entities.

The present application first provides a case handling method. Case handling refers to the handling of a specific matter. For example, in the field of patent agency, patent attorneys need to handle multiple patent application cases, and there are many patent attorneys who can handle various patent application cases. Cases are not well matched with patent attorneys so that patent application cases are not optimally handled. Therefore, in the case of multiple people handling cases and/or multiple cases to be assigned, it is a very difficult problem to well adapt case handling personnel to cases. The case handling method provided in this application can make a good match between case handling personnel and cases. The case handling method provided in this application can be applied to various scenarios where cases need to be handled, for example, it can be used in scenarios where judges, lawyers, patent attorneys, and police are handling cases, and it can also be used in collection scenarios in the financial field.

The implementation terminal of this application can be any device with computing, processing and communication functions, which can be connected to an external device for receiving or sending data, and specifically can be a portable mobile device, such as a smart phone, tablet computer, notebook computer, PDA (Personal Digital Assistant), etc., can also be fixed devices, such as computer equipment, field terminals, desktop computers, servers, workstations, etc., or a collection of multiple devices, such as cloud computing physical infrastructure or server clusters .

Optionally, the implementation terminal of the present application may be a server or a physical infrastructure of cloud computing.

FIG. 1 is a schematic diagram of a system architecture of a case processing method according to an exemplary embodiment. As shown in FIG. 1 , the system architecture includes a server 110 , a user terminal 120 and a database 130 . The user terminal 120 and the server 110 and the database 130 and the server 110 are connected through wired or wireless communication links. Therefore, the user terminal 120 and the database 130 can send data to the server 110, and can also receive data from the server 110. The server 110 is the implementation terminal in this embodiment, and the database 130 stores a sample set. When the case processing method provided by the present application is applied to the system architecture shown in FIG. 1 , a specific process may be as follows: the server 110 first obtains the case sample data in the sample set from the database 130 . For a case, the case The corresponding case sample data includes the attributes and attribute values of the person handling the case, the attributes and attribute values of the case itself, and the case processing result; then, the server 110 uses the case sample data to perform model training to obtain the case success rate model and the case processing respectively. Then, when the user terminal 120 submits a pending case to the server 110, the server 110 can use the combined model to recommend a case handler suitable for handling the pending case, and recommend the The number of case handlers can be well matched with the pending case.

Fig. 2 is a flow chart of a case processing method according to an exemplary embodiment. This embodiment can be executed by a server, as shown in FIG. 2 , and includes the following steps:

Step 230, using the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler, respectively.

The case sample data includes attributes of the case, attribute values corresponding to the attributes of the case, attributes of the case handlers, attribute values corresponding to the attributes of the case handlers, and case handling results.

Attributes of cases and case handlers may also be referred to as features of cases and case handlers, and attribute values corresponding to attributes may also be referred to as feature values corresponding to features. The case processing result can be either a discrete value such as case processing success or failure, or a continuous value such as a case processing score.

In one embodiment, the use of the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler, respectively, includes:

Divide the attribute values corresponding to each attribute of the case and the case handler in the case sample data according to predetermined rules, and obtain the attribute value division interval corresponding to each attribute;

Iteratively execute the target attribute and target attribute value division interval determination step until the number of determined target attributes reaches a second predetermined number, and the target attribute and target attribute value division interval determination step includes: calculating the value of each attribute in the case sample data. Information gain, and select the attribute with the largest information gain from all the unselected attributes as the target attribute, and take the attribute value division interval corresponding to the target attribute as the target attribute value division interval.

In this embodiment, the information gain is used to select the target attribute. Of course, in practical application, the information gain rate or the Gini coefficient can also be used for selection.

For example, in the collection field, the attributes of the collection cases in the case sample data can include the case amount, then the attribute values corresponding to the case amount can be divided into 10,000 or less, 10,000 to 20,000, 20,000 to 30,000, 30,000 to 40,000. For another example, the attributes of the collectors in the case sample data can include age, then the attribute values corresponding to age can be divided into 20-25, 25-30, 30-35, 35-40, 40 and above, etc. Of course, here The age group can also be obtained by dividing the attribute value corresponding to the attribute of the borrower's age in the collection case.

In one embodiment, the predetermined rules include:

For the discrete attributes of the case and the case handler in the case sample data, the attribute value corresponding to each discrete attribute is used as the attribute value corresponding to the attribute to divide the interval;

For the continuous attributes of the case and the case handler in the case sample data, the highest attribute value and the lowest attribute value corresponding to each continuous attribute are equally divided into a third predetermined number of attribute value intervals, which are used as attribute value division intervals corresponding to the attributes.

For example, in the field of collections, the discrete attributes of the collection personnel may include the collection personnel's gender, birthplace, working years, etc. The attribute values corresponding to the collection personnel's gender are "Male" and "Female" respectively, and the collection personnel's working years The corresponding attribute values may be "half a year", "one year", "one and a half years", "two years" and so on.

Discrete attributes of a collection case can include the borrower's gender, place of birth, whether he is single, whether the borrower has been collected before, whether he has other liabilities, and so on.

Still referring to the above example, if the continuous attributes of the collection case in the case sample data can include the case amount, the minimum and maximum attribute values corresponding to the case amount are 10,000 and 60,000 respectively, and the third predetermined number is 5, then the case The attribute value corresponding to the amount can be divided into 10,000 to 20,000, 20,000 to 30,000, 30,000 to 40,000, 40,000 to 50,000, and 50,000 to 60,000. A total of 5 attribute value division intervals.

FIG. 3 is a flowchart of steps before step 230 according to an embodiment shown in the embodiment of FIG. 2 . As shown in FIG. 3, before step 230, it further includes:

Step 210, each time a case handler completes a case, the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handlers, the attribute values corresponding to the attributes of the case handlers, and the cases of the case. The processing result is recorded as the case data corresponding to the case.

When a case handler completes a case, the property values of the case itself, the property values of the case handler, and the case handling result involved in the case will be saved, which can be stored in a database.

In step 220, a sample set is established by using the case data corresponding to all the cases recorded in the history.

In this embodiment, since the sample set is established by using the case data corresponding to all the cases in history, rich data can be provided for establishing the model, thereby ensuring the accuracy and reliability of the established model.

Next, continue to refer to FIG. 2 .

Step 240: For each group of first case sample data in which the target attribute value of the corresponding case handler is in the same target attribute value division interval, use the target attribute of the case and the corresponding target attribute of the case in the first case sample data of the group. The target attribute value and the case processing result are used to train the case success rate model, and the case success rate model corresponding to the first case sample data of the group is obtained.

The case success rate model can be trained by algorithms based on various principles, such as machine learning models such as logistic regression models, neural network models and deep learning models.

The trained case success rate model can recommend case handlers suitable for handling the pending case for the pending case.

For each set of first case sample data, a corresponding case success rate model will be trained.

By using the first case sample data with the target attribute value of a group of corresponding case handlers in the same target attribute value division interval to carry out the corresponding case success rate model, the impurities and interference factors of the data can be reduced, and the accuracy of the trained model can be ensured. sex.

Step 250, for each group of second case sample data in which the target attribute value of the corresponding case is in the same target attribute value division interval, the target attribute of the case handler and the target attribute of the case handler in the second case sample data of the group are used. The corresponding target attribute value and the case processing result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second set of case sample data is obtained.

The success rate model of case handlers can also be trained by algorithms based on various principles, such as machine learning models such as logistic regression models, neural network models and deep learning models.

The trained case handler success rate model can recommend pending cases suitable for the case handler to be handled by the case handler.

Step 260: Combine each of the case success rate models with each of the case handler success rate models, and use the combined model to recommend cases or case handlers, so as to process cases according to the recommendation results.

After combining each case success rate model with each case handler success rate model, the functions of each model are included in the combined model, and each case success rate model and each case handler success rate model are the combined model. Sub-models, specifically, when using the combined model for recommendation, only one or more sub-models in the combined model may be used.

In one embodiment, after combining each of the case success rate models with each of the case handler success rate models, and using the combined model to recommend cases or case handlers, so as to perform case processing according to the recommendation results, The method also includes:

When the case is processed by the case handler, use the case processing result, the attribute value of the case and the attribute value of the case handler to create case sample data, and add the created case sample data to the sample set;

Every predetermined period of time, the case sample data in the sample set is re-trained to train the case success rate model and the case handler success rate model.

In this embodiment, after the new case is processed by the case handler, the model is trained again by using the attribute value of the new case, the attribute value of the corresponding case handler, and the case handling result, thereby effectively ensuring that the model can continue to be Upgrade and update, so that the performance of the model can be continuously improved, and the model can be continuously optimized.

In one embodiment, the use of the combined model to recommend cases or case handlers includes:

When a pending case is received, obtain the target attribute value corresponding to the target attribute of the pending case;

Input the target attribute value corresponding to the target attribute of the case to be processed into each case success rate model in the combined model, and input the target attribute value of each case handler into each of the combined models respectively. The case success rate model, to obtain the success rate of each case handler for the pending case;

According to the success rate of each case handler on the pending case, a case handler suitable for handling the pending case is determined, and the determined case handler is recommended.

Specifically, after inputting the target attribute value corresponding to the target attribute of the case to be processed into each case success rate model in the combined model, first, it is determined that the target attribute value corresponding to the target attribute of the case to be processed matches Then input the target attribute value of each case handler into the determined case success rate model.

Recommendations are made to the identified case handlers, and the recommended content can be displayed by means of pop-up windows, page loading, etc.

In this embodiment, for one or more pending cases, the case success rate model in the combined model is used to implement the recommendation of case handlers who are suitable for handling the pending cases.

In one embodiment, determining a case handler suitable for handling the pending case according to the success rate of each case handler on the pending case includes:

Identifying the case handler with the highest success rate in handling said pending case;

The identified case handler is taken as a case handler suitable for handling the pending case.

Sort each case handler according to the success rate of each case handler on the pending cases from large to small;

One case handler is randomly selected from the first predetermined number of case handlers in the top row as a case handler suitable for handling the pending case.

Determine case handlers whose handling success rate for the pending case is greater than a predetermined handling success rate threshold;

One case handler is randomly selected among the determined case handlers as a case handler suitable for handling the case to be handled.

Because the processing success rate predicted by the model may not fully represent the degree to which a case handler can successfully complete the processing of the pending case, that is, when the first case handler's processing success rate is greater than that of the second case handler, the first case handler The case handler may also handle the case better than the second case handler. In this embodiment, by selecting one of the case handlers whose handling success rate of the case to be handled is greater than the predetermined handling success rate threshold as a case handler suitable for handling the case to be handled, it is possible to ensure that the case to be handled and the case handling While improving the matching degree of personnel, the fairness of selecting case handlers is improved.

In one embodiment, after determining a case handler suitable for handling the pending case according to the success rate of each case handler on the pending case, and recommending the determined case handler, The method also includes:

Record and statistically determine the handling strategies of the case handlers when handling the pending cases;

When the determined case handler is recommended again, determining the busyness of the recommended case handler;

If the busyness is greater than a predetermined busyness threshold, the case handler is re-recommended, and the handling policy is pushed to the re-recommended case handler.

Busyness is an indicator that reflects the pressure of case handlers to handle cases. It can be defined in various ways. For example, it can be defined as the number of pending cases remaining in the day, or it can be defined as the average number of pending cases handled within a certain period of time. .

The handling strategy is the way the case is handled.

For example, in the field of collection, collection strategies can include vocabulary, tone, title, collection method, etc., which can help collection.

In this embodiment, when the busyness is greater than the predetermined busyness threshold, the case handler is re-recommended, so that the case handler is too busy to cause a backlog of cases. The handling strategy of the case handler is pushed to the re-recommended case handler, so that the re-recommended case handler can handle the pending cases well.

When receiving the case handler of the case to be assigned, obtain the target attribute value corresponding to the target attribute of the case handler of the case to be assigned;

Input the target attribute value corresponding to the target attribute of the case handler of the case to be allocated into the success rate model of each case handler in the combined model, and input the target attribute value of each case to be allocated into each case respectively. Calculate the success rate model of each case handler in the combined model, and obtain the handling success rate of each case to be assigned after being handled by the case handler of the case to be assigned;

According to the processing success rate of the cases to be assigned by the case handlers of the cases to be assigned, determine the cases to be assigned that are suitable for being handled by the case handlers of the cases to be assigned. case is recommended.

Specifically, after inputting the target attribute value corresponding to the target attribute of the case handler of the case to be assigned into the success rate model of each case handler in the combined model, first, it is necessary to determine the target attribute corresponding to the case handler. The case handler success rate model that matches the target attribute value of the case handler, and then input the target attribute value of each case to be assigned into the determined case handler success rate model.

In this embodiment, for one or more case handlers, the case handler success rate model in the combined model is used to implement the recommendation of a case suitable for handling by the case handler.

It can be understood that the cases to be assigned and the case handlers of the cases to be assigned are virtual objects represented in a digital manner, and they can have corresponding mappings in the real world.

When applying the case handling method provided by this application to the field of collection, a basic process can be shown in Figure 4 above. Fig. 4 shows a schematic flow chart of the basic flow of the case processing method applied to the collection field according to an exemplary embodiment. Please refer to Figure 4. The specific process is as follows: the collection personnel collect cases, obtain successful cases, save the collection strategies corresponding to these successful cases, and then use the attribute values of these cases to establish the attribute values corresponding to the collection personnel. The case success rate model is based on the case success rate model, and the case success rate model learns the collection case index information that the corresponding collection personnel are good at. When there is a new collection case that needs to be processed, the collection case success rate model determines and assigns the collection personnel suitable for handling the case to be collected according to the attribute value of the new collection case and the attribute value of each collection personnel, and provides the corresponding collection personnel with the corresponding collection personnel. Saved collection strategies, so that for a batch of new collection cases, matching collection personnel can be assigned to them, and these collection personnel can be assisted by collection strategies, so as to efficiently carry out collection work.

To sum up, according to the case handling method provided by the embodiment of FIG. 2, on the one hand, by constructing a case success rate model, a case handler suitable for handling the pending case can be automatically recommended for the pending case; By building a case handler success rate model, the case handler can automatically recommend pending cases suitable for the case handler to handle. Therefore, the case handler can be well matched with the case, thereby improving case handling. Quality and case handling efficiency.

The present application also provides a case processing device, and the following are the device embodiments of the present application.

Fig. 5 is a block diagram of a case processing apparatus according to an exemplary embodiment. As shown in Figure 5, the device 500 includes:

The determination module 510 is configured to use the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler respectively, wherein the case sample data includes the attribute of the case and the attribute value corresponding to the attribute of the case , the attributes of the case handlers, the attribute values corresponding to the attributes of the case handlers, and the case handling results;

The first training module 520 is configured to use the target attribute of the case in the group of the first case sample data, The target attribute value corresponding to the target attribute of the case and the case processing result are used to train the case success rate model, and the case success rate model corresponding to the first case sample data of the group is obtained;

The second training module 530 is configured to use the target attributes of the case handlers in the group of second case sample data, The target attribute value corresponding to the target attribute of the case handler and the case handling result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

The combination and recommendation module 540 is configured to combine each of the case success rate models with each of the case handler success rate models, and use the combined model to recommend cases or case handlers, so as to process cases according to the recommendation results .

In one embodiment, the combining and recommending module is further configured to:

In one embodiment, when the combination and recommendation module determines a case handler suitable for handling the pending case according to the processing success rate of each case handler for the pending case, it is further configured to:

In one embodiment, the determining module is further configured to, before using the case sample data in the sample set to determine the target attribute and the target attribute value division interval for the case and the case handler, respectively:

Whenever a case handler completes a case, the property of the case, the property value corresponding to the property of the case, the property of the case handler, the property value corresponding to the property of the case handler, and the case handling result of the case are taken as Record the case data corresponding to the case;

A sample set is established by using the case data corresponding to all the cases recorded in history.

In one embodiment, the determining module is further configured to:

In one embodiment, the combination and recommendation module is further configured to determine a case handler suitable for handling the pending case according to the success rate of each case handler on the pending case, and to determine the case handler for the pending case. After a referral from the said case handler:

According to a third aspect of the present application, a computer device is also provided, which executes all or part of the steps of any one of the above-mentioned case processing methods. The computer equipment includes:

at least one processor; and

a memory communicatively coupled to the at least one processor; wherein,

The memory stores instructions executable by the at least one processor, the instructions being executed by the at least one processor to enable the at least one processor to execute as illustrated in any of the above-described exemplary embodiments method of handling the case.

As will be appreciated by one skilled in the art, various aspects of the present application may be implemented as a system, method or program product. Therefore, various aspects of the present application can be embodied in the following forms, namely: a complete hardware implementation, a complete software implementation (including firmware, microcode, etc.), or a combination of hardware and software aspects, which may be collectively referred to herein as implementations "circuit", "module" or "system".

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

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

Wherein, the storage unit stores program codes, and the program codes can be executed by the processing unit 610, so that the processing unit 610 executes various exemplary methods according to the present application described in the above-mentioned “Methods of Embodiments” of this specification. Implementation steps.

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) 621 and/or a cache storage unit 622 , and may further include a read only storage unit (ROM) 623 .

The storage unit 620 may also include a program/utility 624 having a set (at least one) of program modules 625 including, but not limited to, an operating system, one or more application programs, other program modules, and program data, An implementation of a network environment may be included in each or some combination of these examples.

The bus 630 may be representative of one or more of several types of bus structures, including a memory cell bus or memory cell controller, a peripheral bus, a graphics acceleration port, a processing unit, or a local area using any of a variety of bus structures bus.

The computer device 600 may also communicate with one or more external devices 800 (eg, keyboards, pointing devices, Bluetooth devices, etc.), with one or more devices that enable a user to interact with the computer device 600, and/or with Any device (eg, router, modem, etc.) that enables the computer device 600 to communicate with one or more other computer devices. Such communication may take place through an input/output (I/O) interface 650 , such as with display unit 640 . Also, the computer device 600 may communicate with one or more networks (eg, a local area network (LAN), a wide area network (WAN), and/or a public network such as the Internet) through a network adapter 660 . As shown, network adapter 660 communicates with other modules of computer device 600 via bus 630 . It should be understood that, although not shown, other hardware and/or software modules may be used in conjunction with computer device 600, including but not limited to: microcode, device drivers, redundant processing units, external disk drive arrays, RAID systems, tape drives and data backup storage systems.

From the description of the above embodiments, those skilled in the art can easily understand that the exemplary embodiments described herein may be implemented by software, or may be implemented by software combined with necessary hardware. Therefore, the technical solutions according to the embodiments of the present application may be embodied in the form of software products, and the software products may be stored in a non-volatile storage medium (which may be CD-ROM, U disk, mobile hard disk, etc.) or on the network , including several instructions to cause a computer device (which may be a personal computer, a server, a terminal device, or a network device, etc.) to execute the method according to the embodiment of the present application.

According to a fourth aspect of the present application, there is also provided a computer-readable storage medium on which a program product capable of implementing the above-mentioned method of the present specification is stored, and the computer-readable storage medium may be non-volatile or easily accessible. loss of sex. In some possible implementations, various aspects of the present application can also be implemented in the form of a program product, which includes program code, which is used to cause the program product to run on a terminal device when the program product is executed. The terminal device performs the steps according to various exemplary embodiments of the present application described in the above-mentioned "Example Method" section of this specification.

Referring to FIG. 7 , a program product 700 for implementing the above method according to an embodiment of the present application is described, which can adopt a portable compact disk read only memory (CD-ROM) and include program codes, and can be used in a terminal device, For example running on a personal computer. However, the program product of the present application is not limited thereto, and in this document, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

The program product may employ any combination of one or more readable media. The readable medium may be a readable signal medium or a readable storage medium. The readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus or device, or a combination of any of the above. More specific examples (non-exhaustive list) of readable storage media include: electrical connections with one or more wires, portable disks, hard disks, random access memory (RAM), read only memory (ROM), erasable programmable read only memory (EPROM or flash memory), optical fiber, portable compact disk read only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

A computer readable signal medium may include a propagated data signal in baseband or as part of a carrier wave with readable program code embodied thereon. Such propagated data signals may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the foregoing. A readable signal medium can also be any readable medium, other than a readable storage medium, that can transmit, propagate, or transport the program for use by or in connection with the instruction execution system, apparatus, or device.

Program code embodied on a readable medium may be transmitted using any suitable medium, including but not limited to wireless, wireline, optical fiber cable, RF, etc., or any suitable combination of the foregoing.

Program code for carrying out the operations of the present application may be written in any combination of one or more programming languages, including object-oriented programming languages—such as Java, C++, etc., as well as conventional procedural Programming Language - such as the "C" language or similar programming language. The program code may execute entirely on the user's computer device, partly on the user's computer device, as a stand-alone software package, partly on the user's computer device and partly on a remote computer device, or entirely on the remote computer device or execute on the server. Where remote computer equipment is involved, the remote computer equipment may be connected to the user computer equipment through any kind of network, including a local area network (LAN) or wide area network (WAN), or may be connected to external computer equipment (eg, using an Internet service provider business via an Internet connection).

In addition, the above-mentioned figures are only schematic illustrations of the processes included in the methods according to the exemplary embodiments of the present application, and are not intended to be limiting. It is easy to understand that the processes shown in the above figures do not indicate or limit the chronological order of these processes. In addition, it is also readily understood that these processes may be performed synchronously or asynchronously, for example, in multiple modules.

It should be understood that the present application is not limited to the precise structures described above and illustrated in the accompanying drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the application is limited only by the appended claims.

Claims

A case handling method that includes:

Use the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler respectively, wherein the case sample data includes the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handler, The attribute value corresponding to the attribute of the case handler and the case handling result;

For the first case sample data in which the target attribute value of each group of corresponding case handlers is in the same target attribute value division interval, the target attribute value of the case and the target attribute value corresponding to the target attribute of the case in the first case sample data of the group are used. and the case processing results to train the case success rate model to obtain the case success rate model corresponding to the sample data of the first case in the group;

For the second case sample data in which the target attribute value of each group of corresponding cases is in the same target attribute value division interval, the target attribute of the case handler and the target attribute corresponding to the target attribute of the case handler in the second case sample data of the group are used. The attribute value and the case processing result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

Each of the case success rate models and each of the case handler success rate models are combined, and the combined model is used to recommend a case or a case handler, so as to handle the case according to the recommendation result.
The method according to claim 1, wherein the recommendation of cases or case handlers by using the combined model comprises:

When a pending case is received, obtain the target attribute value corresponding to the target attribute of the pending case;

Input the target attribute value corresponding to the target attribute of the case to be processed into each case success rate model in the combined model, and input the target attribute value of each case handler into each of the combined models respectively. The case success rate model, to obtain the success rate of each case handler for the pending case;

According to the success rate of each case handler on the pending case, a case handler suitable for handling the pending case is determined, and the determined case handler is recommended.
The method according to claim 2, wherein determining a case handler suitable for handling the to-be-handled case according to the success rate of each case-handling staff in handling the to-be-handled case comprises:

Sort each case handler according to the success rate of each case handler on the pending cases from large to small;

One case handler is randomly selected from the first predetermined number of case handlers in the top row as a case handler suitable for handling the pending case.
The method according to claim 1, wherein the recommendation of cases or case handlers by using the combined model comprises:

When receiving the case handler of the case to be assigned, obtain the target attribute value corresponding to the target attribute of the case handler of the case to be assigned;

Input the target attribute value corresponding to the target attribute of the case handler of the case to be allocated into the success rate model of each case handler in the combined model, and input the target attribute value of each case to be allocated into each case respectively. Calculate the success rate model of each case handler in the combined model, and obtain the handling success rate of each case to be assigned after being handled by the case handler of the case to be assigned;

According to the processing success rate of the cases to be assigned by the case handlers of the cases to be assigned, determine the cases to be assigned that are suitable for being handled by the case handlers of the cases to be assigned. case is recommended.
The method according to claim 1, wherein before using the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler respectively, the method further comprises:

Whenever a case handler completes a case, the property of the case, the property value corresponding to the property of the case, the property of the case handler, the property value corresponding to the property of the case handler, and the case handling result of the case are taken as Record the case data corresponding to the case;

A sample set is established by using the case data corresponding to all the cases recorded in history.
The method according to claim 1, wherein said using the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler respectively, comprising:

Divide the attribute values corresponding to each attribute of the case and the case handler in the case sample data according to predetermined rules, and obtain the attribute value division interval corresponding to each attribute;

Iteratively execute the target attribute and target attribute value division interval determination step until the number of determined target attributes reaches a second predetermined number, and the target attribute and target attribute value division interval determination step includes: calculating the value of each attribute in the case sample data. Information gain, and select the attribute with the largest information gain from all the unselected attributes as the target attribute, and take the attribute value division interval corresponding to the target attribute as the target attribute value division interval.
The method according to claim 2, wherein, according to the processing success rate of each case handling person on the pending case, a case handling person suitable for handling the pending case is determined, and the determined case is processed. After the person recommends, the method further includes:

Record and statistically determine the handling strategies of the case handlers when handling the pending cases;

When the determined case handler is recommended again, determining the busyness of the recommended case handler;

If the busyness is greater than a predetermined busyness threshold, the case handler is re-recommended, and the handling policy is pushed to the re-recommended case handler.
A case processing device, comprising:

The determining module is configured to use the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler respectively, wherein the case sample data includes the attribute of the case, the attribute value corresponding to the attribute of the case, Attributes of case handlers, attribute values corresponding to the attributes of case handlers, and case handling results;

The first training module is configured to use the target attribute of the case, the case in the group of the first case sample data, and the The target attribute value corresponding to the target attribute and the case processing result are used to train the case success rate model, and the case success rate model corresponding to the first case sample data of the group is obtained;

The second training module is configured to use the target attributes of the case handlers, the The target attribute value corresponding to the target attribute of the handler and the case handling result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

The combination and recommendation module is configured to combine each of the case success rate models with each of the case handler success rate models, and use the combined model to recommend cases or case handlers, so as to process cases according to the recommendation results.
A computer device, comprising a memory and a processor, wherein computer-readable instructions are stored in the memory, and when the computer-readable instructions are executed by the processor, the processor is caused to execute:

Use the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler respectively, wherein the case sample data includes the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handler, The attribute value corresponding to the attribute of the case handler and the case handling result;

For the first case sample data in which the target attribute value of each group of corresponding case handlers is in the same target attribute value division interval, the target attribute value of the case and the target attribute value corresponding to the target attribute of the case in the first case sample data of the group are used. and the case processing results to train the case success rate model to obtain the case success rate model corresponding to the sample data of the first case in the group;

For the second case sample data in which the target attribute value of each group of corresponding cases is in the same target attribute value division interval, the target attribute of the case handler and the target attribute corresponding to the target attribute of the case handler in the second case sample data of the group are used. The attribute value and the case processing result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

Each of the case success rate models and each of the case handler success rate models are combined, and the combined model is used to recommend a case or a case handler, so as to handle the case according to the recommendation result.
The computer device according to claim 9, wherein the recommendation of cases or case handlers by using the combined model comprises:

When a pending case is received, obtain the target attribute value corresponding to the target attribute of the pending case;

Input the target attribute value corresponding to the target attribute of the case to be processed into each case success rate model in the combined model, and input the target attribute value of each case handler into each of the combined models respectively. The case success rate model, to obtain the success rate of each case handler for the pending case;

According to the success rate of each case handler on the pending case, a case handler suitable for handling the pending case is determined, and the determined case handler is recommended.
The computer device according to claim 10, wherein the determining a case handler suitable for handling the pending case according to the success rate of each case handler on the pending case, comprises:

Sort each case handler according to the success rate of each case handler on the pending cases from large to small;

One case handler is randomly selected from the first predetermined number of case handlers in the top row as a case handler suitable for handling the pending case.
The computer device according to claim 9, wherein the recommendation of cases or case handlers by using the combined model comprises:

When receiving the case handler of the case to be assigned, obtain the target attribute value corresponding to the target attribute of the case handler of the case to be assigned;

Input the target attribute value corresponding to the target attribute of the case handler of the case to be allocated into the success rate model of each case handler in the combined model, and input the target attribute value of each case to be allocated into each case respectively. Calculate the success rate model of each case handler in the combined model, and obtain the handling success rate of each case to be assigned after being handled by the case handler of the case to be assigned;

According to the processing success rate of the cases to be assigned by the case handlers of the cases to be assigned, determine the cases to be assigned that are suitable for being handled by the case handlers of the cases to be assigned. case is recommended.
10. The computer device of claim 9, wherein the computer-readable instructions are executed by the processor before using the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler, respectively , causing the processor to also execute:

Whenever a case handler completes a case, the property of the case, the property value corresponding to the property of the case, the property of the case handler, the property value corresponding to the property of the case handler, and the case handling result of the case are taken as Record the case data corresponding to the case;

A sample set is established by using the case data corresponding to all the cases recorded in history.
The computer device according to claim 9, wherein said using the case sample data in the sample set to determine the target attribute and the target attribute value division interval for the case and the case handler respectively, comprising:

Divide the attribute values corresponding to each attribute of the case and the case handler in the case sample data according to predetermined rules, and obtain the attribute value division interval corresponding to each attribute;

Iteratively execute the target attribute and target attribute value division interval determination step until the number of determined target attributes reaches a second predetermined number, and the target attribute and target attribute value division interval determination step includes: calculating the value of each attribute in the case sample data. Information gain, and select the attribute with the largest information gain from all the unselected attributes as the target attribute, and take the attribute value division interval corresponding to the target attribute as the target attribute value division interval.
The computer device according to claim 10, wherein, according to the processing success rate of each case handler for the pending case, a case handler suitable for handling the pending case is determined, and the determined case is After the processor makes the recommendation, when the computer-readable instructions are executed by the processor, the processor further executes:

Record and statistically determine the handling strategies of the case handlers when handling the pending cases;

When the determined case handler is recommended again, determining the busyness of the recommended case handler;

If the busyness is greater than a predetermined busyness threshold, the case handler is re-recommended, and the handling policy is pushed to the re-recommended case handler.
A computer-readable storage medium storing computer-readable instructions that, when executed by one or more processors, cause the one or more processors to execute:

Use the case sample data in the sample set to determine target attributes and target attribute value division intervals for the case and the case handler respectively, wherein the case sample data includes the attributes of the case, the attribute values corresponding to the attributes of the case, the attributes of the case handler, The attribute value corresponding to the attribute of the case handler and the case handling result;

For the first case sample data in which the target attribute value of each group of corresponding case handlers is in the same target attribute value division interval, the target attribute value of the case and the target attribute value corresponding to the target attribute of the case in the first case sample data of the group are used. and the case processing results to train the case success rate model to obtain the case success rate model corresponding to the sample data of the first case in the group;

For the second case sample data in which the target attribute value of each group of corresponding cases is in the same target attribute value division interval, the target attribute of the case handler and the target attribute corresponding to the target attribute of the case handler in the second case sample data of the group are used. The attribute value and the case processing result are used to train the case handler success rate model, and the case handler success rate model corresponding to the second case sample data of the group is obtained;

Each of the case success rate models and each of the case handler success rate models are combined, and the combined model is used to recommend a case or a case handler, so as to handle the case according to the recommendation result.
The computer-readable storage medium according to claim 16, wherein the recommendation of cases or case handlers by using the combined model comprises:

When a pending case is received, obtain the target attribute value corresponding to the target attribute of the pending case;

Input the target attribute value corresponding to the target attribute of the case to be processed into each case success rate model in the combined model, and input the target attribute value of each case handler into each of the combined models respectively. The case success rate model, to obtain the success rate of each case handler for the pending case;

According to the success rate of each case handler on the pending case, a case handler suitable for handling the pending case is determined, and the determined case handler is recommended.
The computer-readable storage medium according to claim 17, wherein the determining a case handler suitable for handling the to-be-handled case according to the processing success rate of each case-handling person on the to-be-handled case comprises:

Sort each case handler according to the success rate of each case handler on the pending cases from large to small;

One case handler is randomly selected from the first predetermined number of case handlers in the top row as a case handler suitable for handling the pending case.
The computer-readable storage medium according to claim 16, wherein the recommendation of cases or case handlers by using the combined model comprises:

When receiving the case handler of the case to be assigned, obtain the target attribute value corresponding to the target attribute of the case handler of the case to be assigned;

Input the target attribute value corresponding to the target attribute of the case handler of the case to be allocated into the success rate model of each case handler in the combined model, and input the target attribute value of each case to be allocated into each case respectively. Calculate the success rate model of each case handler in the combined model, and obtain the handling success rate of each case to be assigned after being handled by the case handler of the case to be assigned;

According to the processing success rate of the cases to be assigned by the case handlers of the cases to be assigned, determine the cases to be assigned that are suitable for being handled by the case handlers of the cases to be assigned. case is recommended.
17. The computer-readable storage medium of claim 16, wherein before using the case sample data in the sample set to determine the target attribute and target attribute value division interval for the case and the case handler, respectively, the computer-readable instructions are executed by one or more When multiple processors execute, cause one or more processors to also execute:

Whenever a case handler completes a case, the property of the case, the property value corresponding to the property of the case, the property of the case handler, the property value corresponding to the property of the case handler, and the case handling result of the case are taken as Record the case data corresponding to the case;

A sample set is established by using the case data corresponding to all the cases recorded in history.