Incoming call processing method, electronic device and computer readable storage medium

Abstract

The present invention provides a method for processing incoming calls, an electronic device, and a computer-readable storage medium. In the method, in response to receiving an incoming call from a user, user information associated with the incoming call number is obtained, and then the user information is input into a first tree model, and the prediction result output by the first tree model is received; wherein the first tree model is used to predict the question that the user wants to consult, and the first tree model is composed of a plurality of branch models, and the prediction accuracy of any one of the branch models is greater than or equal to a preset accuracy threshold, thereby connecting the user's incoming call to a target customer service number, and the target customer service number is associated with the prediction result. The technical solution provided by the present invention can improve the prediction accuracy of the customer service system for user calls and improve the efficiency of incoming call processing.

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to an incoming call processing method, an electronic device and a computer-readable storage medium. Background Art

[0002] With the popularity of online transactions, intelligent customer service systems are increasingly involved in people's lives. Users can call customer service to consult questions. Customer service systems generally work in a mode where users can select consultation scenarios on their own. That is, users can select the business scenario corresponding to the question they want to consult by pressing a button. Then, the customer service system transfers the user's call to the manual customer service corresponding to the business scenario selected by the user.

[0003] However, the accuracy of users' self-selected business scenarios is low, generally only about 15%. This further leads to the problem of mismatch between user consultation questions and customer service skills. Usually, manual customer service is required to communicate with users, and after determining the specific questions of users, manual customer service will manually transfer the user's call to the manual customer service corresponding to the actual business scenario. This call processing method is cumbersome, users have to wait for a long time, and the accuracy and efficiency are low. Summary of the invention

[0004] The present invention provides a method for processing incoming calls, an electronic device and a computer-readable storage medium, which are used to improve the prediction accuracy of a customer service system for user calls and improve the efficiency of processing incoming calls.

[0005] In a first aspect, the present invention provides a method for processing an incoming call, comprising:

[0006] In response to receiving an incoming call from a user, obtaining user information associated with the incoming call number;

[0007] Inputting the user information into a first tree model, and receiving a prediction result output by the first tree model; wherein the first tree model is used to predict the question to be consulted by the user, the first tree model is composed of a plurality of branch models, and the prediction accuracy of any one of the branch models is greater than or equal to a preset accuracy threshold;

[0008] The user's incoming call is connected to a target customer service number, and the target customer service number is associated with the prediction result.

[0009] In a second aspect, the present invention provides an electronic device, comprising:

[0010] An acquisition module, configured to acquire user information associated with an incoming call number in response to receiving an incoming call from a user;

[0011] A processing module, used to input the user information into a first tree model, and receive a prediction result output by the first tree model; wherein the first tree model is used to predict the question to be consulted by the user, the first tree model is composed of a plurality of branch models, and the prediction accuracy of any one of the branch models is greater than or equal to a preset accuracy threshold;

[0012] A switching module is used to connect the user's incoming call to a target customer service number, and the target customer service number is associated with the prediction result.

[0013] In a third aspect, the present invention provides an electronic device, comprising: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the incoming call processing method described in the first aspect are implemented.

[0014] In a fourth aspect, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein the computer program is executed by a processor to implement the method as described in the first aspect.

[0015] The present invention provides a method for processing incoming calls, an electronic device and a computer-readable storage medium. When a user call is received, user information associated with the incoming call number is obtained, and then a first tree model is used to predict the question the user wants to ask, and then the user's call is automatically forwarded. There is no need for the user to independently select a business scenario, which simplifies the user's operation steps and is conducive to improving processing efficiency. Moreover, in an embodiment of the present invention, the first tree model is composed of a plurality of branch models with high prediction accuracy, and the prediction accuracy is high. The tree model has high interpretability, which is convenient for customer service personnel to understand the user situation and further improve the efficiency of incoming call processing. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the present disclosure and, together with the description, serve to explain the principles of the present disclosure.

[0017] Figure 1 A schematic diagram of a business scenario of a method for handling incoming calls provided by an embodiment of the present invention;

[0018] Figure 2 A flowchart of a method for processing incoming calls provided by an embodiment of the present invention;

[0019] Figure 3 A schematic diagram of the structure of a first tree model provided by an embodiment of the present invention;

[0020] Figure 4 A schematic diagram of a flow chart of a training method for a first tree model provided in an embodiment of the present invention;

[0021] Figure 5 Another structural schematic diagram of the first tree model provided by an embodiment of the present invention;

[0022] Figure 6 A flowchart of another incoming call processing method provided by an embodiment of the present invention;

[0023] Figure 7 A functional block diagram of an electronic device provided by an embodiment of the present invention;

[0024] Figure 8 A schematic diagram of the physical structure of an electronic device provided by an embodiment of the present invention.

[0025] The above drawings show clear embodiments of the present disclosure, which will be described in more detail below. These drawings and text descriptions are not intended to limit the scope of the present disclosure in any way, but to illustrate the concepts of the present disclosure to those skilled in the art by referring to specific embodiments. DETAILED DESCRIPTION

[0026] Exemplary embodiments will be described in detail herein, examples of which are shown in the accompanying drawings. When the following description refers to the drawings, the same numbers in different drawings represent the same or similar elements unless otherwise indicated. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with the present disclosure. Instead, they are merely examples of devices and methods consistent with some aspects of the present disclosure as detailed in the appended claims.

[0027] The incoming call processing method provided by the embodiment of the present invention can be applied to Figure 1 The business scenario diagram shown in Figure 1 is as follows. Figure 1 As shown, in this business scenario, the customer service system can communicate with one or more user terminals ( Figure 1 3 user terminals are shown) for communication, and the user calls made by the user through the terminal devices such as mobile phones are transferred to the manual customer service.

[0028] Among them, the terminal device can be a wireless terminal or a wired terminal. The wireless terminal can be a device that provides voice and / or other business data connectivity to users, a handheld device with wireless connection function, or other processing equipment connected to a wireless modem. The wireless terminal can communicate with one or more core network devices via a radio access network (Radio Access Network, referred to as RAN). The wireless terminal can be a mobile terminal, such as a mobile phone (or "cellular" phone) and a computer with a mobile terminal. For example, it can be a portable, pocket-sized, handheld, computer-built-in or vehicle-mounted mobile device that exchanges language and / or data with the radio access network. For another example, the wireless terminal can also be a personal communication service (Personal Communication Service, referred to as PCS) phone, a cordless phone, a Session Initiation Protocol (Session Initiation Protocol, referred to as SIP) phone, a wireless local loop (Wireless Local Loop, referred to as WLL) station, a personal digital assistant (Personal Digital Assistant, referred to as PDA) and other devices. The wireless terminal may also be referred to as a system, a subscriber unit, a subscriber station, a mobile station, a mobile station, a remote station, a remote terminal, an access terminal, a user terminal, a user agent, a user device or user equipment, without limitation. Optionally, the terminal device may also be a smart watch, a tablet computer or other device. The terminal device has the function of making calls through a network or an operator.

[0029] The specific application scenario of the present invention is the processing scenario of a customer service system for a user's incoming call.

[0030] For example, in one possible implementation scenario, after a user places an order through a shopping platform, if he needs further consultation on logistics issues or after-sales issues of the order, he can call the customer service phone number. At this time, the customer service system of the shopping platform (or merchant) needs to transfer the user's call to the customer service personnel corresponding to a specific business.

[0031] For another example, in another possible implementation scenario, the user may also call the customer service telephone number of the communication operator to inquire about or handle communication services.

[0032] For another example, in another possible implementation scenario, the user can also call the customer service phone number of the application (Application, APP) to contact the customer service staff of the APP by phone, and then consult problems in using the APP or handle business.

[0033] In the disclosed embodiments, for any customer service system, when it receives a user call, it is generally necessary to determine the question the user wants to consult, so as to assign the call to the corresponding customer service personnel according to the business field to which the question belongs. However, as described in the background technology, when the user independently selects the business scenario, the operation steps are cumbersome and the accuracy is low.

[0034] For example, a user wants to consult about logistics issues, but because he is not clear about the business division method in the customer service system, he chooses the business scenario of after-sales consultation. At this time, the customer service system will transfer the user's call to the customer service staff responsible for after-sales consultation. However, the customer service staff responsible for after-sales consultation may not be clear about the problems in the logistics business and cannot answer the user's questions. In other words, there is a mismatch between the customer service staff's skills and the user's consultation questions.

[0035] In this case, customer service personnel can be trained in all skills, but the training cycle is long and the business scope is wide, which makes it difficult for customer service personnel to digest and the training effect is poor. Alternatively, the customer service personnel need to communicate with the user by phone first, and then manually transfer the call after clarifying the actual question asked by the user. For example, in the previous example, the customer service personnel responsible for after-sales consultation need to transfer the user's call to the customer service personnel responsible for logistics business after understanding the actual question asked by the user. This processing method prolongs the overall call time, and may also require a second transfer, and the call processing efficiency is poor.

[0036] The technical solution provided by the present invention is intended to solve the above technical problems of the prior art.

[0037] The following specific embodiments are used to describe in detail the technical solutions of the present invention and how the technical solutions of the present application solve the above-mentioned technical problems. The following specific embodiments can be combined with each other, and the same or similar concepts or processes may not be repeated in some embodiments. The embodiments of the present invention will be described below in conjunction with the accompanying drawings.

[0038] Embodiment 1

[0039] The embodiment of the present invention provides a method for processing incoming calls. Figure 1 The customer service system shown. Below, for ease of understanding, the customer service system of the shopping platform is taken as an example to illustrate the call processing method.

[0040] Please refer to Figure 2The incoming call processing method provided by the embodiment of the present invention may include the following steps:

[0041] S202: In response to receiving an incoming call from a user, obtain user information associated with the incoming call number.

[0042] In the embodiment of the present invention, user information may include but is not limited to: user data includes at least one of user basic information, user behavior information, and order information.

[0043] The basic user information may include, but is not limited to, one or more of: user name, user account (e.g., user code of a shopping platform), user contact information (e.g., mobile phone number, contact email address, etc.), user age and gender. The aforementioned basic user information may include user registration information.

[0044] User behavior information may include, but is not limited to: one or more of the user's purchase behavior information, evaluation behavior information, click behavior information, and consultation behavior information on the shopping platform.

[0045] The order information may include, but is not limited to, at least one of: basic order information, order logistics status information, order timeliness information, and transaction product information (for example, product type, product price, product store information, etc.).

[0046] As before, this step is used to obtain user information related to the incoming call number, and the user information can characterize the user from at least one of the above aspects. In the embodiment of the present invention, the obtained user information can be user information of one or more users, and the acquisition method will be described in detail later in conjunction with the specific implementation method.

[0047] In a possible embodiment of the present invention, user data is data updated in real time by a Kafka system. In this way, after a user calls in, feature information of the second-level delay before the call is received can be obtained, which helps to combine real-time data to further improve the accuracy of the prediction results, and further improve the matching degree between user consultation questions and customer service skills, thereby improving user experience.

[0048] S204, inputting user information into the first tree model, and receiving the prediction result output by the first tree model; wherein the first tree model is used to predict the question that the user wants to ask, the first tree model is composed of a plurality of branch models, and the prediction accuracy of any branch model is greater than or equal to a preset accuracy threshold.

[0049] The first tree model may be a Boosted Trees model. The first tree model may also be called a problem prediction model, a problem prejudgment model or other names, and the embodiment of the present invention has no particular limitation on the name.

[0050] Please refer to Figure 3 , Figure 3 A schematic diagram of the structure of the first tree model is shown. Figure 3 As shown, the first tree model is composed of multiple branch models in a tree structure, and each branch model also includes a root node (the node at the top from top to bottom) and multiple leaf nodes (the nodes at the bottom from top to bottom). Each leaf node is a question that a user may ask. Based on different user information input into the first tree model, different prediction results (questions that the user may ask) can be output according to the tree structure of the first tree model.

[0051] In such Figure 3 In the first tree model shown, any root node to a leaf node below can be used as a prediction logic; and a branch model can be used as a prediction logic set. In a specific implementation scenario, multiple prediction logic sets of each branch model in the first tree model can be stored in a readable storage location of the customer service system. For example, it can be stored in the local memory of the customer service system, or in the memory of another electronic device connected or communicating with the customer service system.

[0052] Moreover, in the embodiment of the present invention, in the first tree model, the prediction accuracy of each branch model can reach or exceed the accuracy threshold, which can ensure that the accuracy of the prediction result is high to a certain extent. Among them, for any leaf node, the prediction accuracy of the leaf node can be the proportion of the number of positive samples in the total number of samples, wherein the positive sample refers to the sample whose prediction result is the same as the actual result and the prediction is accurate. In addition, there may also be negative samples in the sample, which are samples other than the positive samples in the total samples, and the prediction result is different from the actual result.

[0053] The embodiment of the present invention uses a first tree model with a high prediction accuracy to achieve problem prediction. While ensuring a high accuracy of the output result, the output result can also be explained based on the tree structure, with high interpretability and high stability of problem prediction.

[0054] The construction and training of the first tree model will be described in detail later and will not be repeated here.

[0055] S206, connecting the user's incoming call to a target customer service number, and the target customer service number is associated with the prediction result.

[0056] At this time, after predicting the questions that the user may ask based on the first tree model, the business department to which the user belongs can be determined based on the questions that the user may ask, and then the user's call can be transferred to a target customer service number of the business department to enable the user to talk with the customer service staff.

[0057] In such Figure 2 In the illustrated embodiment, the user does not need to customize the business scenario of the question to be consulted. Instead, the customer service system automatically outputs the prediction result and then automatically transfers the call by processing the user information using the first tree model. For the caller, this processing method simplifies the user operation, can effectively shorten the waiting time before accessing the manual customer service, and provides a high user experience. Moreover, in the embodiment of the present invention, the first tree model is composed of a plurality of branch models with high prediction accuracy, and the prediction accuracy is high. The tree model has high interpretability, which makes it easier for customer service personnel to understand the user situation and further improves the efficiency of call processing.

[0058] Below, in Figure 2 Based on the above, the incoming call processing method provided by the embodiment of the present invention is further described.

[0059] In the embodiment of the present invention, the first tree model involved in S204 may be obtained by training with sample data, that is, firstly, sample data is obtained, and then the first tree model is trained with the sample data.

[0060] Specifically, the method of training the first tree model can refer to Figure 4 The flow chart of the model training method shown in Figure 4 As shown, the first tree model can be trained as follows.

[0061] S402, obtaining sample data.

[0062] The sample data contains at least two types of data: user information and user consultation questions. The user information is the same as before and will not be repeated here. The user consultation questions are used to determine the prediction accuracy of the model. For any sample data, the yoghurt information is input into the model. If the prediction result output by the model is the same as the user consultation question, the sample data is a positive sample.

[0063] In addition, the "prediction result is the same as the user's inquiry question" mentioned here means that the two have the same semantics, not that all the words are the same. For example, if the prediction result is "query the current logistics status", and the user's inquiry question in the sample data is "where is the logistics now", the semantics of the two are the same, then the prediction result is the same as the user's inquiry question.

[0064] In a specific implementation, the sample data may be offline data of a shopping platform. The offline data may come from a big data set of the shopping platform. By analyzing the data in the big data set, user information and user consultation questions are obtained, and then the user information and user consultation questions are used to build a model and verify the data accuracy.

[0065] Among them, the big data set can be built based on hive, python, and spark development. When analyzing the data in the big data set, appropriate features can be selected through business induction and data analysis. These features describe the status of the customer's phone call, mainly including order information, order logistics status, order promise timeliness, consulting behavior, traffic click behavior, etc. Through vectorization processing (or numerical processing), sample data is obtained, and then the sample data is used for the training of the first tree model.

[0066] S404, using the sample data to train the initial tree model to obtain a second tree model; wherein the second tree model is fully fitted, the input data of the second tree model is the sample user data in the sample data, and the output data is the predicted questions asked by the user.

[0067] The initial tree model may be a preset tree model having a root node and a plurality of leaf nodes. In the embodiment of the present invention, the sample data is used to train a second tree model that is sufficiently fitted.

[0068] S406, split the second tree model into multiple branch models; the branch models include a first branch model and a second branch model, the prediction accuracy of the first branch model is greater than or equal to the accuracy threshold, and the prediction accuracy of the second branch model is less than the accuracy threshold.

[0069] Specifically, starting from the root node, from top to bottom, the second tree model can be pruned into multiple branch models, each branch model having its own root node and leaf nodes. When pruning, the number of first branch models can be one or more, and the number of second branch models can be one or more, which is determined by the prediction accuracy of each branch model.

[0070] S408, using the sample data, training the second branch model to obtain a third branch model, wherein the prediction accuracy of the third branch model is greater than or equal to the accuracy threshold.

[0071] This step is used to train the second branch model with a lower prediction accuracy. This step can be repeated multiple times until a third branch model with a higher prediction accuracy is obtained.

[0072] S410, combining the first branch model and the third branch model to obtain a first tree model.

[0073] That is, the first branch model and the third branch model are packaged to obtain the first tree model.

[0074] When executing this step, the related model logics can be combined into a tree structure, and the unrelated branch models can be directly packaged. At this time, the first tree model can be expressed as one or more tree structures. For example, Figure 3 The first tree model shown is a tree structure. For another example, Figure 5 Another structural schematic diagram of the first tree model is shown. As shown in FIG5 , the first tree model is composed of three tree structures, and a tree structure can be composed of one or more branch models.

[0075] As before, the first tree model has a high degree of interpretability. Figure 3 The first tree model is used as an example for explanation. When the processing logic in the first tree model is extracted, the following prediction logic can be obtained:

[0076] When X≤1, the prediction result is question A1;

[0077] When X>1 and Y!=3 (!= indicates not equal), the prediction result is question A2;

[0078] When X>1, Y=3 and Z>5, the prediction result is question A3;

[0079] When X>1, Y=3, Z≤5 and Q<6, the prediction result is question A4;

[0080] When X>1, Y=3, Z≤5 and Q≥6, the prediction result is question A5.

[0081] Therefore, based on the prediction result output by the first tree model, the prediction result can be explained according to the above prediction logic. For example, if the user information obtained from the user call is input into the first tree model, and the prediction result output by the first tree model is question A3, it can be determined that the prediction result is obtained by satisfying the condition "X>1, Y=3 and Z>5".

[0082] In a possible embodiment, in addition to forwarding the user's call in the manner shown in the above embodiment, the customer service system can also output the first tree model to the customer service node corresponding to the target customer service number. In this way, the customer service staff can understand the user information of the user's call based on the first tree model, which is conducive to subsequent efficient communication.

[0083] In a possible design of an embodiment of the present invention, an association relationship between each incoming call number and user information can be directly established, so that when the customer service system receives an incoming call from a user, it can directly determine the user information associated with the incoming call number based on the association relationship.

[0084] In another possible design, the user information can be associated with the user account, that is, the user information is stored according to the user account. Thus, the customer service system first determines the user account corresponding to the incoming call number.

[0085] In this case, a number set can also be maintained in another readable storage location of the customer service system, such as in an HBase database. The number set can also be called a phone dimension table, which is used to associate and maintain the user account of the user on the shopping platform with the phone number. Multiple phone number pools can be maintained in the phone dimension table, and one phone number pool corresponds to one user account. In the phone number pool of any user account, multiple phone numbers can be associated and maintained. These phone numbers can be numbers that have behavioral associations with the user account.

[0086] Specifically, for any user account, multiple phone numbers associated with the user account can be collected; then, the multiple phone numbers are stored as a number set of the user account. For example, in one possible embodiment, the phone numbers of the recipients in the most recent M orders of the user account, the phone number used for the most recent N customer service consultations of the user account, and the phone number left when the user account was registered can be obtained for the user account to which the orders of the shopping platform belong within the past year, and these M+N+1 phone numbers are associated with the user account and stored to form a phone number pool of the user account. Among them, M and N are both integers greater than 0, and the two can be the same or different. The embodiment of the present invention has no special limitation on this value. For example, M can be 5 and N can be 3. In this way, each user account can be associated with 9 phone numbers.

[0087] In addition, in another possible implementation scenario, the numbers in the telephone maintenance table are encrypted, which can improve the information security of the customer service system to a certain extent.

[0088] Based on this design, when the customer service system executes the aforementioned step S202, when receiving a user call, it determines the target number set to which the incoming call number belongs, thereby determining the user account corresponding to the target number set, and then obtaining the user information corresponding to the user account.

[0089] At this time, due to the multiple phone numbers maintained in the phone maintenance table, there may be a phone number that may fall into multiple phone number pools, that is, there are multiple target number sets to which the incoming call number belongs. At this time, when the customer service system executes the subsequent S204 step, the user information corresponding to each user account can be input into the first tree model respectively, and multiple prediction results output by the first tree model can be received respectively, each prediction result corresponding to a user account.

[0090] After receiving the prediction result output by the first tree model, the customer service system also needs to determine the associated target customer service number based on the prediction result.

[0091] In a possible embodiment, another readable storage location of the customer service system may also store a correspondence table between the prediction results (or questions asked) and the customer service number. In this way, after receiving the prediction results, the calibration system can determine the target customer service number based on the correspondence table.

[0092] In another possible embodiment, a correspondence between the prediction results (or questions asked) and the business departments may be established, so that the manual customer service telephone number in each business department may be used as the target customer service number.

[0093] In an embodiment of the present invention, the first tree model may output multiple prediction results. For example, when the acquired user information is associated with multiple user accounts, the prediction results associated with each user account output by the first tree model may be received. For another example, even if the acquired user information is associated with only one user account, the first tree model may output multiple prediction results with a high prediction accuracy.

[0094] When there are multiple prediction results, the customer service system can determine the target prediction result from the multiple prediction results, and the prediction accuracy of the branch model to which the target prediction result belongs is the highest among the multiple prediction results (that is, the prediction result with the highest accuracy can be obtained as the target prediction result); thereby, the target business department to which the target prediction result belongs is determined, and then the user's call is connected to the target customer service number of the target business department. That is, the customer service system can obtain the prediction result with the highest prediction accuracy to transfer the user's call.

[0095] In another embodiment, when there are multiple prediction results, the customer service phone number of the business department with the most prediction results can be obtained according to the business departments corresponding to the prediction results, and used as the target customer service number. For example, if there are 5 prediction results, of which 2 prediction results correspond to business department 1, 3 prediction results correspond to business department 2, and business department 2 has the largest number of prediction results, the customer service phone number of business department 2 is used as the target customer service number.

[0096] In actual scenarios, you can also determine only one target business department, and through the business interface, transfer the user's call to the customer service allocation system of the business department, and then let the customer service allocation system of the business department automatically allocate manual customer service. That is, the connection number of the manual customer service assigned by the target business department is used as the target customer service number.

[0097] Through the above process, after the user dials the customer service number, the customer service system answers the user's call and transfers the user's call to the target customer service number according to the above process, so that the user can communicate with the manual customer service by phone. In a possible embodiment, the user number can also be encrypted for the customer service staff to ensure the security of user information.

[0098] Figure 6 FIG. 2 shows another flow chart of the incoming call processing method provided by an embodiment of the present invention. Figure 6 As shown, the customer service system can use offline data, such as order details, logistics information, user basic information, etc., to build a first tree model; and for real-time data on the shopping platform, it is synchronized and stored in real time through the Kafka system. In addition, the customer service system also maintains a telephone dimension table. Therefore, when the customer service system receives a user call, it queries the telephone dimension table through the business interface and the data interface to determine the target user account, and then obtains the user data synchronized in real time by the Kafka system to obtain the user information. After that, the customer service system can input the user information into the first tree model through the data interface and receive the prediction results output by it. Based on the predicted questions that the user may ask, the target business department can be determined. Furthermore, after the customer service system obtains the target business department through the business interface, it can transfer the user's call to the manual customer service responsible for the business department to realize the conversation between the user and the manual customer service.

[0099] It is to be understood that some or all of the steps or operations in the above embodiments are merely examples, and the present application embodiments may also perform other operations or variations of various operations. In addition, the various steps may be performed in different orders presented in the above embodiments, and it is possible that not all of the operations in the above embodiments need to be performed.

[0100] The words used in this application are only used to describe the embodiments and are not used to limit the claims. As used in the description of the embodiments and claims, unless the context clearly indicates, the singular forms of "a", "an" and "the" are intended to include the plural forms as well. Similarly, the term "and / or" as used in this application refers to any and all possible combinations of one or more associated listings. In addition, when used in this application, the term "comprise" and its variants "comprises" and / or including (comprising) refer to the existence of stated features, wholes, steps, operations, elements, and / or components, but do not exclude the existence or addition of one or more other features, wholes, steps, operations, elements, components and / or groups of these.

[0101] Embodiment 2

[0102] Based on the incoming call processing method provided in the first embodiment above, the present invention further provides a device embodiment for implementing each step and method in the above method embodiment.

[0103] An embodiment of the present invention provides an electronic device, please refer to Figure 7 , the electronic device 700 comprises:

[0104] An acquisition module 72 is used to acquire user information associated with the incoming call number in response to receiving an incoming call from a user;

[0105] The processing module 74 is used to input the user information into the first tree model and receive the prediction result output by the first tree model; wherein the first tree model is used to predict the question to be consulted by the user, and the first tree model is composed of a plurality of branch models, and the prediction accuracy of any branch model is greater than or equal to a preset accuracy threshold;

[0106] The transfer module 76 is used to connect the user's call to the target customer service number, and the target customer service number is associated with the prediction result.

[0107] In one embodiment of the present invention, the electronic device 700 further includes a model training module ( Figure 7 (not shown), the model training module is used to:

[0108] Get sample data;

[0109] The first tree model is trained using the sample data.

[0110] Specifically, the model training module is used to:

[0111] The initial tree model is trained using the sample data to obtain a second tree model; wherein the second tree model is fully fitted, the input data of the second tree model is the sample user data in the sample data, and the output data is the predicted question asked by the user;

[0112] The second tree model is divided into a plurality of branch models; the branch models include a first branch model and a second branch model, the prediction accuracy of the first branch model is greater than or equal to the accuracy threshold, and the prediction accuracy of the second branch model is less than the accuracy threshold;

[0113] Using the sample data, the second branch model is trained to obtain a third branch model, wherein the prediction accuracy of the third branch model is greater than or equal to the accuracy threshold;

[0114] The first branch model and the third branch model are combined to obtain a first tree model.

[0115] In another embodiment of the present invention, the acquisition module 72 is specifically configured to:

[0116] In response to receiving an incoming call from a user, determining a target number set to which the incoming call number belongs;

[0117] Determine the user account corresponding to the target number set;

[0118] Get the user information corresponding to the user account.

[0119] In another embodiment, when there are multiple user accounts, the processing module 74 is specifically configured to:

[0120] Inputting user information corresponding to each user account into the first tree model respectively;

[0121] A plurality of prediction results output by the first tree model are received, each prediction result corresponding to a user account.

[0122] In one embodiment of the present invention, the electronic device 700 further includes:

[0123] Acquisition module ( Figure 7 (not shown), for collecting, for any user account, a plurality of telephone numbers associated with the user account;

[0124] Storage Module( Figure 7 (not shown) for storing multiple telephone numbers as a number set of user accounts.

[0125] In one embodiment of the present invention, the switching module 76 is specifically used for:

[0126] Determine a target prediction result among multiple prediction results, where the prediction accuracy of the branch model to which the target prediction result belongs is the highest among the multiple prediction results;

[0127] Determine the target business unit to which the target forecast results belong;

[0128] Connect the user's call to the target customer service number of the target business department.

[0129] In the embodiment of the present invention, the user data includes at least one of user basic information, user behavior information, and order information.

[0130] In one embodiment of the present invention, user data is data updated in real time by the Kafka system.

[0131] In one embodiment of the present invention, the electronic device 700 further includes:

[0132] Output Module( Figure 7 (not shown) is used to output the first tree model to the customer service node corresponding to the target customer service number.

[0133] Figure 7The electronic device 700 of the illustrated embodiment can be used to execute the technical solution of the above-mentioned method embodiment. Its implementation principle and technical effects can be further referred to the relevant description in the method embodiment. Optionally, the electronic device 700 can be a server or a terminal.

[0134] It should be understood that the above Figure 7 The division of the modules of the electronic device 700 shown is only a division of logical functions. In actual implementation, they can be fully or partially integrated into one physical entity, or they can be physically separated. And these modules can all be implemented in the form of software calling through processing elements; they can also be all implemented in the form of hardware; some modules can also be implemented in the form of software calling through processing elements, and some modules can be implemented in the form of hardware. For example, the processing module 74 can be a separately established processing element, or it can be integrated in the electronic device 700, such as a chip of the terminal. In addition, it can also be stored in the memory of the electronic device 700 in the form of a program, and called and executed by a processing element of the electronic device 700. The implementation of other modules is similar. In addition, these modules can be fully or partially integrated together, or they can be implemented independently. The processing element here can be an integrated circuit with signal processing capabilities. In the implementation process, each step of the above method or each module above can be completed by an integrated logic circuit of hardware in the processor element or instructions in the form of software.

[0135] For example, the above modules may be one or more integrated circuits configured to implement the above methods, such as one or more application specific integrated circuits (ASIC), or one or more digital singnal processors (DSP), or one or more field programmable gate arrays (FPGA). For another example, when a certain module above is implemented in the form of a processing element scheduler, the processing element may be a general-purpose processor, such as a central processing unit (CPU) or other processor that can call a program. For another example, these modules may be integrated together and implemented in the form of a system-on-a-chip (SOC).

[0136] Furthermore, an embodiment of the present invention provides an electronic device, please refer to Figure 8 , the electronic device 700 comprises:

[0137] Memory 710;

[0138] Processor 720; and

[0139] Computer programs;

[0140] The computer program is stored in the memory 710 and is configured to be executed by the processor 720 to implement the method of the above embodiment.

[0141] The number of processors 720 in the electronic device 700 may be one or more, and the processor 720 may also be referred to as a processing unit, which may implement certain control functions. The processor 720 may be a general-purpose processor or a dedicated processor, etc. In an optional design, the processor 720 may also store instructions, which may be executed by the processor 720, so that the electronic device 700 executes the method described in the above method embodiment.

[0142] In yet another possible design, the electronic device 700 may include a circuit that can implement the functions of sending, receiving, or communicating in the aforementioned method embodiments.

[0143] Optionally, the number of memories 710 in the electronic device 700 may be one or more, and instructions or intermediate data are stored in the memory 710, and the instructions can be executed on the processor 720, so that the electronic device 700 performs the method described in the above method embodiment. Optionally, other relevant data may also be stored in the memory 710. Optionally, instructions and / or data may also be stored in the processor 720. The processor 720 and the memory 710 may be provided separately or integrated together.

[0144] In addition, if Figure 8 As shown, a transceiver 730 is also provided in the electronic device 700, wherein the transceiver 730 can be called a transceiver unit, a transceiver, a transceiver circuit, or a transceiver, etc., and is used for data transmission or communication with a test device or other terminal device, which will not be repeated here.

[0145] like Figure 8 As shown, the memory 710, the processor 720 and the transceiver 730 are connected and communicated via a bus.

[0146] In addition, an embodiment of the present invention provides a readable storage medium on which a computer program is stored. The computer program is executed by a processor to implement the method described in the first embodiment.

[0147] Since each module in this embodiment can execute the method shown in the first embodiment, for the parts not described in detail in this embodiment, reference may be made to the relevant description of the first embodiment.

[0148] Those skilled in the art will readily appreciate other embodiments of the present disclosure after considering the specification and practicing the invention disclosed herein. The present invention is intended to cover any variations, uses or adaptations of the present disclosure that follow the general principles of the present disclosure and include common knowledge or customary techniques in the art that are not disclosed in the present disclosure. The description and examples are to be regarded as exemplary only, and the true scope and spirit of the present disclosure are indicated by the following claims.

[0149] It should be understood that the present disclosure is not limited to the exact structures that have been described above and shown in the drawings, and that various modifications and changes may be made without departing from the scope thereof. The scope of the present disclosure is limited only by the appended claims.

Claims

1. A method for handling incoming calls, characterized in that: include: In response to receiving an incoming call from a user, obtaining user information associated with the incoming call number; The user information is input into a first tree model, and a prediction result output by the first tree model is received; wherein the first tree model is used to predict the question that the user wants to consult, and the first tree model is composed of a plurality of branch models, and the prediction accuracy of any one of the branch models is greater than or equal to a preset accuracy threshold; each branch model serves as a prediction logic set, and the prediction logic set includes at least one prediction logic, and the prediction logic is from any root node to a leaf node below; the prediction logic is used to explain the prediction result; the user's call is connected to a target customer service number, and the target customer service number is associated with the prediction result.

2. The method according to claim 1, characterized in that The method further comprises: Get sample data; The first tree model is trained using the sample data.

3. The method according to claim 2, characterized in that The step of training the first tree model using the sample data includes: The sample data is used to train the initial tree model to obtain a second tree model; wherein the second tree model is fully fitted, the input data of the second tree model is the sample user data in the sample data, and the output data is the predicted question asked by the user; The second tree model is divided into a plurality of branch models; the branch models include a first branch model and a second branch model, the prediction accuracy of the first branch model is greater than or equal to the accuracy threshold, and the prediction accuracy of the second branch model is less than the accuracy threshold; Using the sample data, training the second branch model to obtain a third branch model, wherein the prediction accuracy of the third branch model is greater than or equal to the accuracy threshold; The first branch model and the third branch model are combined to obtain the first tree model.

4. The method according to any one of claims 1 to 3, characterized in that: The step of obtaining user information associated with the incoming call number in response to receiving an incoming call from a user includes: In response to receiving an incoming call from a user, determining a target number set to which the incoming call number belongs; Determine the user account corresponding to the target number set; The user information corresponding to the user account is obtained.

5. The method according to claim 4, characterized in that When the number of the user accounts is multiple, inputting the user information into the first tree model and receiving the prediction result output by the first tree model includes: Inputting the user information corresponding to each of the user accounts into the first tree model respectively; Receive a plurality of prediction results output by the first tree model, each prediction result corresponding to one of the user accounts.

6. The method according to claim 4, characterized in that The method further comprises: For any one of the user accounts, collecting a plurality of telephone numbers associated with the user account; The multiple telephone numbers are stored as a number set of the user account.

7. The method according to any one of claims 1 to 3, characterized in that: When the number of the prediction results is multiple, the step of connecting the user's incoming call to a target customer service number includes: Determine a target prediction result among the multiple prediction results, wherein the prediction accuracy of the branch model to which the target prediction result belongs is the highest among the multiple prediction results; Determine the target business department to which the target forecast result belongs; Connect the user's incoming call to the target customer service number of the target business department.

8. The method according to any one of claims 1 to 3, characterized in that: The user data includes at least one of user basic information, user behavior information, and order information.

9. The method according to claim 8, characterized in that The user data is updated in real time by the Kafka system.

10. The method according to any one of claims 1 to 3, characterized in that: The method further comprises: The first tree model is output to the customer service node corresponding to the target customer service number.

11. An electronic device, characterized in that: include: An acquisition module, configured to acquire user information associated with an incoming call number in response to receiving an incoming call from a user; A processing module, used for inputting the user information into a first tree model, and receiving a prediction result output by the first tree model; wherein the first tree model is used for predicting the question to be consulted by the user, the first tree model is composed of a plurality of branch models, and the prediction accuracy of any one of the branch models is greater than or equal to a preset accuracy threshold; each branch model is used as a prediction logic set, the prediction logic set includes at least one prediction logic, and the prediction logic is from any root node to a leaf node below; the prediction logic is used for explaining the prediction result; A switching module is used to connect the user's incoming call to a target customer service number, and the target customer service number is associated with the prediction result.

12. An electronic device, characterized in that: The method comprises a memory, a processor and a computer program stored in the memory and executable on the processor, wherein when the processor executes the program, the steps of the incoming call processing method according to any one of claims 1 to 10 are implemented.

13. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the incoming call processing method according to any one of claims 1 to 10 are implemented.

14. A computer program product, comprising a computer program, wherein when the computer program is executed by a processor, the computer program implements the steps of the incoming call processing method according to any one of claims 1 to 10.

Images