CN113507541A

CN113507541A - Incoming call intention display method and device

Info

Publication number: CN113507541A
Application number: CN202110751704.2A
Authority: CN
Inventors: 李健; 郭剑霓; 王素文; 吴海英; 蒋宁; 曾琳铖曦
Original assignee: Mashang Consumer Finance Co Ltd
Current assignee: Mashang Consumer Finance Co Ltd
Priority date: 2021-07-02
Filing date: 2021-07-02
Publication date: 2021-10-15

Abstract

The application provides a method and a device for displaying incoming call intentions, wherein the method for displaying the incoming call intentions comprises the following steps: basic information of the incoming call user is displayed in a first display area of an operation interface, and the basic information is obtained through matching of an incoming call identifier of the incoming call user; before or when the incoming call is connected to be successful, displaying an intention label of the incoming call user in a second display area of the operation interface, wherein the intention label is used for representing the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on basic information or incoming call identification matching. Therefore, when the user calls, the customer service staff can determine the intention of the user at the moment through the intention label of the incoming call user displayed in the second display area, so that the customer service can determine the intention of the incoming call user more quickly and accurately, the communication efficiency of the customer service and the incoming call user is improved, and better service is provided for the incoming call user.

Description

Incoming call intention display method and device

Technical Field

The present application relates to the field of computer technologies, and in particular, to a method and an apparatus for displaying an incoming call intention.

Background

The call intention refers to the purpose of a user to dial a customer service hotline. When a user has certain requirements to be solved, the user usually dials a customer service hot line, and the requirements are solved by customer service staff in the customer service hot line. And the customer service staff need to provide corresponding services for the user according to the incoming call intention of the user.

Currently, the customer service staff determines the incoming call intention of the incoming call user mainly through the following two ways. The first mode is as follows: after the customer service staff establishes connection with the incoming call user, the customer service staff confirms the incoming call intention of the incoming call user through communication with the incoming call user. The second mode is as follows: after the incoming call user calls, the customer service staff presumes the incoming call intention of the incoming call user according to the work order information, the telephone summary and other information displayed on the workbench.

However, the first method for determining the incoming call intention of the incoming call user needs to rely on the communication between the customer service staff and the incoming call user. Different customer service personnel have different communication abilities, and some customer service personnel cannot accurately and quickly determine the incoming call intention of the incoming call user. And under the condition that the incoming call of the incoming call user is complained, the incoming call user sometimes sends out the mismatching of the heart, and at the moment, customer service staff are more difficult to know the real incoming call intention of the incoming call user through communication with the incoming call user. The second method is adopted to determine the incoming call intention of the incoming call user, and the incoming call intention needs to depend on the information such as the work order information, the telephone summary and the like displayed on the workbench. The information is recorded by the customer service staff who is connected with the telephone line, and if the previous customer service staff do not record well, the customer service staff who is connected with the telephone line sees the information which is scattered and incomplete, so that the customer service staff cannot accurately determine the incoming call intention of the incoming call user.

Disclosure of Invention

The embodiment of the application aims to provide a method and a device for displaying an incoming call intention so as to improve the accuracy of customer service in determining the intention of an incoming call user.

In order to solve the above technical problem, an embodiment of the present application provides the following technical solutions:

a first aspect of the present application provides a method for displaying an incoming call intention, where the method includes: basic information of a calling user is displayed in a first display area of an operation interface, and the basic information is obtained through matching of a calling identification of the calling user; before or when the incoming call is connected to be successful, displaying an intention label of the incoming call user in a second display area of the operation interface, wherein the intention label is used for representing the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

A second aspect of the present application provides an incoming call intention display device, the device comprising: the first display module is used for displaying basic information of the incoming call user in a first display area of an operation interface, and the basic information is obtained by matching incoming call identification of the incoming call user; the second display module is used for displaying an intention label of the incoming call user in a second display area of the operation interface before the incoming call is connected to be successful or when the incoming call is connected to be successful, wherein the intention label is used for representing the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

A third aspect of the present application provides an electronic device comprising: a processor, a memory, a bus; the processor and the memory complete mutual communication through the bus; the processor is for invoking program instructions in the memory for performing the method of the first aspect.

A fourth aspect of the present application provides a computer-readable storage medium comprising: a stored program; wherein the program, when executed, controls an apparatus in which the storage medium is located to perform the method of the first aspect.

Compared with the prior art, the incoming call intention display method provided by the first aspect of the application displays the basic information of the incoming call user in the first display area of the operation interface, displays the intention label of the incoming call user in the second display area of the operation interface before the incoming call connection is successful or when the incoming call connection is successful, the intention label is used for representing the purpose of the incoming call user, and is obtained by extracting the behavior data of the incoming call user, so that when the user calls, a customer service staff can determine the identity of the incoming call user through the basic information of the incoming call user displayed in the first display area, and determine the intention of the incoming call user at the moment through the intention label of the incoming call user displayed in the second display area, and the intention label displayed to the customer service is automatically determined based on the behavior data of the incoming call user without the customer service and then determining the intention of the incoming call user according to personal experience, the intention label can represent the historical behaviors of the incoming call user to the greatest extent, so that the customer service can determine the intention of the incoming call user more quickly and accurately, reduce the communication time when the customer service determines the intention of the incoming call user, improve the communication efficiency between the customer service and the incoming call user, meet the appeal of the incoming call user more quickly, and provide higher-quality service for the incoming call user.

The device for displaying the incoming call intention, the electronic device provided by the third aspect and the computer-readable storage medium provided by the fourth aspect have the same or similar advantages as the method for displaying the incoming call intention provided by the first aspect.

Drawings

The above and other objects, features and advantages of exemplary embodiments of the present application will become readily apparent from the following detailed description read in conjunction with the accompanying drawings. Several embodiments of the present application are illustrated by way of example and not by way of limitation in the figures of the accompanying drawings and in which like reference numerals refer to similar or corresponding parts and in which:

fig. 1 is a first flowchart illustrating a method for displaying an incoming call intention according to an embodiment of the present application;

FIG. 2 is a first schematic diagram of an operation interface in an embodiment of the present application;

FIG. 3 is a second flowchart illustrating a method for displaying an incoming call intention according to an embodiment of the present application;

FIG. 4 is a second schematic diagram of an operation interface in the embodiment of the present application;

FIG. 5 is a third schematic view of an operation interface in an embodiment of the present application;

FIG. 6 is a fourth schematic view of an operation interface in the embodiment of the present application;

FIG. 7 is a fifth schematic view of an operation interface in an embodiment of the present application;

FIG. 8 is a sixth schematic view of an operation interface in an embodiment of the present application;

FIG. 9 is a seventh schematic view of an operation interface in the embodiment of the present application;

FIG. 10 is a third flowchart illustrating a method for displaying an incoming call intention according to an embodiment of the present application;

fig. 11 is a schematic flowchart of a method for acquiring behavior data of an incoming call user in an embodiment of the present application;

FIG. 12 is a block diagram of an architecture for storing behavioral data in a data warehouse according to an embodiment of the present disclosure;

FIG. 13 is a flowchart illustrating a method for creating an intention tag library according to an embodiment of the present application;

fig. 14 is a fourth flowchart illustrating a method for displaying an incoming call intention according to an embodiment of the present application;

FIG. 15 is a schematic diagram illustrating an architecture of a method for displaying an incoming call intention according to an embodiment of the present application;

FIG. 16 is a first schematic structural diagram of an incoming call intention display device according to an embodiment of the present application;

FIG. 17 is a second schematic structural diagram of a device for displaying an incoming call intention in an embodiment of the present application;

FIG. 18 is a third schematic structural view of a device for displaying an incoming call intention in an embodiment of the present application;

fig. 19 is a schematic structural diagram of an electronic device in an embodiment of the present application.

Detailed Description

Exemplary embodiments of the present application will be described in more detail below with reference to the accompanying drawings. While exemplary embodiments of the present application are shown in the drawings, it should be understood that the present application may be embodied in various forms and should not be limited to the embodiments set forth herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete, and will fully convey the scope of the disclosure to those skilled in the art.

It is to be noted that, unless otherwise specified, technical or scientific terms used herein shall have the ordinary meaning as understood by those skilled in the art to which this application belongs.

In the prior art, in order to determine the intention of an incoming call user and further provide better service for the incoming call user, the intention of the incoming call user is determined by communicating with the incoming call user, or the intention of the incoming call user is analyzed by checking information displayed on a workbench.

However, the customer service needs to rely on the service ability of the customer service staff whether to communicate with the caller or to determine the intention of the caller by viewing the information displayed on the workbench. Different customer service capacities are different, and information displayed by a workbench is not complete, so that the efficiency and accuracy of determining the intention of the incoming call user are different. That is, in the prior art, the customer service is required to determine the intention of the calling user, and the accuracy and efficiency of determining the intention of the calling user are poor.

It should be noted that the incoming call not only refers to the user calling the customer service through a telephone, but also refers to the user calling the customer service through other communication methods such as an online chat window.

In view of the problems of low accuracy and low efficiency in determining the intention of the incoming call user in the prior art, the embodiment of the application provides an incoming call intention display method, after the user calls and before the customer service connects the incoming call, the incoming call intention labels of the user are generated according to behavior data and the like of the incoming call user, and the intention labels are displayed on a client interface of the customer service. By the method provided by the embodiment, the customer service staff can predict the incoming call intention of the user before the incoming call is connected, and prepare the communication content according to the incoming call intention, so that the communication time can be greatly shortened, the communication efficiency can be improved, and the experience of the customer service and the incoming call user can be further improved.

It should be noted here that the main execution body of the incoming call intention display method provided in the embodiment of the present application is a system or a client used by a customer service.

Fig. 1 is a first schematic flow chart of a method for displaying an incoming call intention in an embodiment of the present application, and referring to fig. 1, the method may include:

s101: basic information of a calling user is displayed in a first display area of an operation interface.

The basic information of the incoming call user is obtained by matching the incoming call identification of the incoming call user. That is, different incoming call users have different incoming call identifications and basic information. The incoming call identification of the same incoming call user and the basic information of the same incoming call user are correspondingly stored, and after the incoming call identification of the incoming call user is obtained, the basic information of the incoming call user can be matched according to the incoming call identification.

For example, the incoming call id of the user a is 13612341234, and the basic information corresponding to the incoming call id 13612341234 is users of zhang san, man, and samsung. The incoming call identifier of the user B is 13800010002, and the basic information corresponding to the incoming call identifier 13800010002 is Liqu, Men and Wuxing users. After the user dials the hotline, the system can obtain the incoming call identification 13612341234 of the incoming call user, and can obtain basic information of Zhangsan, Men and Samsung users of the incoming call user through the incoming call identification, and further display the basic information in a first display area of an operation interface, so that the client can determine the identity of the current incoming call user.

The above-mentioned incoming call identifier is only an example, and the incoming call identifier may also be other identifiers capable of distinguishing different users. The specific content of the incoming call identifier is not limited herein.

The basic information of the incoming call user can be desensitized during display, for example, the basic information of the user a is a page XX, an incoming call number 136XXXX1234, and an identification number 1234.

S102: before or when the incoming call is connected to be successful, the intention label of the incoming call user is displayed in a second display area of the operation interface.

The intention label of the incoming call user is used for representing the purpose of the incoming call user. The intention label of the incoming call user is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on basic information or incoming call identification matching.

The purpose of incoming calls varies from user to user. However, before the user calls, some problems are necessarily encountered before the user calls for help. The behavior data of the incoming call user can be used for knowing what the incoming call user does before the incoming call, and the purpose of the incoming call user can be determined according to the content of the incoming call user.

For example, suppose that user A is browsing a detail page of a XX financial product in the XX application, there are still some questions about the financial product. Then, after the user dials the customer service hotline, the system obtains behavior data of the incoming call user through the incoming call identification of the incoming call user, namely, "browse detail page of XX financial product in XX application", and determines that the incoming call of the incoming call user aims at consulting the financial product by analyzing the behavior data. Further, when the incoming call user is successfully connected to the customer service, the intention label "consult XX financial products" of the user is displayed in the second display area. Therefore, after the user dials the hotline, before the user is successfully connected with the customer service or when the user is successfully connected with the customer service, the customer service can know the incoming call of the user before the customer service communicates with the incoming call user, and the purpose of inquiring XX financial products is to be achieved. The customer service can call up related introduction of the XX financial product in advance so as to answer the calling user.

Fig. 2 is a schematic diagram of an operation interface in an embodiment of the present application, and as shown in fig. 2, in the operation interface, a first display area 201 displays basic information (name, gender, mobile phone number, identity card number, and the like) of a current incoming call user, and a second display area 202 displays intention labels (a consultation repayment mode, a consultation consumption transition period, a consultation penalty collection rule, a consultation credit report rule, a consultation and urging interaction information trueness … of a person outside the category, an expected repayment transaction, a suspension of transaction, and the like) of the current incoming call user. The operation interface is an interface of a system or a client used by a customer service. The customer service can acquire the identity of the calling user through the first display area 201 in the display interface. The customer service can know the purpose of the incoming call of the user at this time through the second display area 202 in the display interface.

As can be seen from the above, in the incoming call intention displaying method provided in the embodiment of the present application, basic information of a caller is displayed in a first display area of an operation interface, an intention tag of the caller is displayed in a second display area of the operation interface before or when the caller is connected successfully, the intention tag is used for representing the purpose of the incoming call of the caller, and is obtained by extracting behavior data of the caller, so that when the user makes an incoming call, a servicer can determine the identity of the caller through the basic information of the caller displayed in the first display area, and determine the intention of the caller at that time through the intention tag of the caller displayed in the second display area, since the intention tag displayed to the servicer is automatically determined based on the behavior data of the caller, the servicer does not need to determine the intention of the caller according to personal experience, the intention label can represent the historical behaviors of the incoming call user to the greatest extent, so that the customer service can determine the intention of the incoming call user more quickly and accurately, reduce the communication time when the customer service determines the intention of the incoming call user, improve the communication efficiency between the customer service and the incoming call user, meet the appeal of the incoming call user more quickly, and provide higher-quality service for the incoming call user.

Further, as a refinement and an extension of the method shown in fig. 1, the embodiment of the present application further provides an incoming call intention display method. Fig. 3 is a second flowchart illustrating a method for displaying an incoming call intention in an embodiment of the present application, and referring to fig. 3, the method may include:

s301: basic information of a calling user is displayed in a first display area of an operation interface.

Step S301 is the same as step S101, and is not described herein again.

S302: before or when the incoming call is connected to be successful, the intention label of the incoming call user is displayed in a second display area of the operation interface.

Step S302 is the same as step S102, and is not described herein again.

S303: and displaying the dialect corresponding to the intention label in a third display area of the operation interface before or when the incoming call is connected to be successful.

Wherein the third display region corresponds to the intent tag of the second display region.

Different intent tags correspond to different dialogs (different dialog patterns for different questions). In order to enable the customer service to solve the problem for the incoming call user more quickly and professionally, the incoming call intention of the incoming call user is displayed in the second display area, simultaneously, the dialogs corresponding to the intention labels in the second display area are also displayed in the third display area beside the second display area, and the dialogs in the third display area correspond to the intention labels in the second display area. Therefore, when the customer service determines that the intention of the incoming call user is one intention label in the second display area, the customer service can directly see the corresponding word operation of the intention label, and can further serve the incoming call user more quickly and professionally.

Even the same intention label may correspond to a plurality of different dialogs. For example: the intention label "consult financial product a" may correspond to a variety of grammatical "product profile", "purchasing restrictions", "purchasing risk", etc. Therefore, it is necessary to correspond each intention label in the second display area with its corresponding plurality of dialogs. For example: the intention label a corresponds to the dialects 1 and 2, and the intention label B corresponds to the dialects 3 and 4.

Fig. 4 is a second schematic diagram of an operation interface in the embodiment of the present application, and referring to fig. 4, on the basis of fig. 2, the operation interface further includes a third display area 203. In the third display area 203, words corresponding to the intention labels in the second display area 202 are displayed (the payment methods of the loan corresponding to the intention label "consultation payment method" include an equal-amount principal payment method, an equal-amount principal payment method ", and the like). The customer service can quickly provide the answer corresponding to the intention label for the incoming call user by looking at the content in the third display area 203.

S304: after the incoming call is successfully connected, the ordering of the intent tags in the second display area is adjusted, or a portion of the intent tags in the second display area are highlighted.

The sequence of the intention labels and the highlighted part of the intention labels are related to the actual purpose of the calling user. The actual purpose of the incoming call user is obtained through the current communication content of the incoming call user.

That is, in the process of the user of the incoming call communicating with the customer service, the intention label ranked at the top (or ranked at the first) in the second display area is not the purpose of the user's incoming call, and the purpose of the user's incoming call may be the intention label ranked at the bottom (or ranked at the second) in the second display area. In order to display the intention label capable of representing the purpose of the user for the incoming call in front so that the customer service can see the intention label at the first time, each intention label in the second display area can be continuously and dynamically adjusted in the communication process according to the actual purpose of the user for the incoming call.

While the actual purpose of the e-subscriber can be obtained through the content of his current communication to the customer service. Specifically, first, the communication content between the incoming call user and the customer service is obtained. For example: the user of the incoming call says what words to the customer service, what words the user of the incoming call inputs, etc. Then, semantic recognition is carried out on the communication content to obtain the actual purpose of the incoming call user. Then, calculating the similarity between each intention label in the second display area and the actual purpose; and finally, the intention labels in the second display area are rearranged according to the sequence of the similarity from high to low. Or, one or more intention labels with the highest similarity to the actual purpose are highlighted (for example, enlarged, thickened, changed in color, and the like) in the second display area.

Fig. 5 is a third schematic view of an operation interface in an embodiment of the application, referring to fig. 5, based on fig. 4, in a process of communication between a customer service and an incoming call user, it is found that the sequenced first or first intention labels in the second display area 202 do not conform to the current communication content of the incoming call user through the communication content of the incoming call user, and then the sequencing of the intention labels in the second display area 202 is adjusted, or a part of the intention labels in the second display area 202 is highlighted (either one of them may be selected, or both of them may be used at the same time). Therefore, the customer service can find the correct intention label more quickly, and further the communication efficiency is improved.

S305: after the incoming call is connected successfully, the sequence of the dialects corresponding to the intention labels is adjusted.

Wherein the ordering of the dialogs corresponding to each intention label is related to the actual dialogs of the customer service. The actual telephone operation of the customer service is obtained through the current communication content of the customer service and the incoming call user.

An intent tag may correspond to multiple utterances that are also arranged in sequence. After the customer service locks an intention tag, the first-ranking corresponding to the intention tag may not be needed by the customer service. In order to enable the customer service to more quickly check the required dialogs, the plurality of dialogs corresponding to the intention labels can be reordered according to the current communication content between the customer service and the incoming call user.

Specifically, firstly, acquiring communication contents of a customer service and an incoming call user; then, calculating the similarity of a plurality of dialogues corresponding to the communication content and the intention label; finally, the multiple utterances are reordered in order of high to low similarity.

Here, the ranking of the dialects may be performed for a plurality of dialects corresponding to all the intention tags in the second display area, or may be performed for a plurality of dialects corresponding to one intention tag selected by a customer in the second display area. The multiple dialogs corresponding to a certain intention label selected by the customer service in the second display area are sequenced, so that the useless dialogs of the intention label can be prevented from being sequenced, the sequencing time is shortened, the customer service can see the latest dialog sequencing result as soon as possible, and the customer service can find out the target dialogs at the first time to serve the calling user conveniently.

Fig. 6 is a fourth schematic view of an operation interface in an embodiment of the present application, and referring to fig. 6, on the basis of fig. 4, in a process of communicating between a customer service and an incoming call user, it is found that, of multiple dialogs corresponding to a certain intention tag in the third display area 203 through communication contents of the customer service and the incoming call user, a first or next-to-front dialogs in the sequence does not correspond to the current communication content of the incoming call user, and then the sequence of the dialogs corresponding to the intention tag in the third display area 203 is adjusted. Therefore, the customer service can find out the real needed words more quickly, and further the communication efficiency is improved.

S306: and after the incoming call is connected successfully, displaying the phrase in the current communication content of the incoming call user in a fourth display area of the operation interface.

When the customer service communicates with the incoming call user, especially under the condition of voice communication through the telephone, if the words of the incoming call user carry dialects or the incoming call user has repeated statements about the same meaning, the customer service cannot effectively communicate with the incoming call user. In view of this, in order to improve the communication efficiency between the client and the incoming call user, corresponding phrases may be extracted from the words spoken by the incoming call client or the input text, and these phrases are displayed in the fourth display area of the operation interface. Therefore, the customer service can clearly understand the meaning which the caller wants to express through the content displayed in the fourth display area, and further the communication efficiency is improved.

Here, the phrase displayed in the fourth display area is a phrase after the duplication is removed.

Fig. 7 is a fifth schematic view of an operation interface in the embodiment of the present application, and referring to fig. 7, a fourth display area 204 is added to the operation interface on the basis of fig. 4. In the fourth display area 204, phrases in the content during the incoming call process of the incoming call user are displayed, such as: consultations, payments, modes, interest rates, calculations, etc. The customer service can clearly understand the meaning that the incoming call user wants to express through each phrase in the fourth display area 204, and further provide corresponding service for the incoming call user in a targeted manner.

S307: and after the incoming call is connected successfully, displaying words in the phrase which do not appear in the intention label of the second display area in a fifth display area of the operation interface.

And an editing window is arranged in the fifth display area. And the customer service can edit the words in the fifth display area through the editing window to generate a new intention label.

In the communication between the customer service and the incoming user, the incoming user may suddenly think of another problem, and the problem is suddenly think of by the incoming user who has not performed the operation related to the problem before. In order to improve the recommendation range of the subsequent intention labels, words which are not included in the intention labels in the second display area can be selected from the phrases displayed in the fourth display area and displayed in the fifth display area. In addition, an editing window for customer service to edit is further provided in the fifth display area, so that the customer service can edit the words displayed in the fifth display area through the editing window, for example: increase "what" before the word "benchmark interest rate", modify "fixed interest" to "fixed interest rate", etc. And the words in the fifth display area are perfected through customer service, a new intention label is generated, and the recommendation range of subsequent intention labels can be improved.

Fig. 8 is a sixth schematic view of an operation interface in an embodiment of the present application, and referring to fig. 8, a fifth display area 205 is added to the operation interface on the basis of fig. 7. In the fifth display area 205, words that are not referred to in the intention labels of the second display area 202 in the phrase of the fourth display area 204 are shown, such as: interest rate, calculation, etc. And there is also an edit window 2051 for the customer service to edit the words therein, thereby generating a new intention label and increasing the recommendation range of subsequent intention labels.

S308: and after the incoming call is connected successfully, displaying the professional terms corresponding to the words in the phrase which do not appear in the intention label of the second display area in a sixth display area of the operation interface.

In the communication between the customer service and the incoming user, the incoming user may suddenly think of another problem, and the problem is suddenly think of by the incoming user who has not performed the operation related to the problem before. That is, a certain vocabulary spoken by the calling user is professional and has no relevant content on the current operation interface, or the certain vocabulary spoken by the calling user may not be a vocabulary in the field. In order to enable the customer service to have feedback on all the questions of the incoming call user, the professional terms corresponding to the words without any association relationship with the intention labels of the second display area in the phrases displayed in the fourth display area may be displayed in the sixth display area.

The term "professional" as used herein may refer to a term that a word is called in the professional field, or may refer to an interpretation that a word is interpreted in the professional field. By displaying the calling of the vocabulary of the caller in the professional field in the sixth display area, the customer service can conveniently and quickly find related contents according to the professional calling, answer the user and improve the service efficiency of the customer service.

Fig. 9 is a seventh schematic view of an operation interface in the embodiment of the present application, and referring to fig. 9, a sixth display area 206 is added to the operation interface on the basis of fig. 8. In the sixth display area 206, there are shown terms corresponding to the words that are not referred to in the intention labels of the second display area 202 in the phrase of the fourth display area 204, such as: interest is principal x shelf life x interest rate, etc. In this way, the customer service can be facilitated to answer the user according to the professional terminology.

It should be noted here that the above steps S302 and S303 may be executed simultaneously, and the steps S304, S305, S306, S307, and S308 may be executed simultaneously. And one or more of steps S303, S304, S305, S306, S307, and S308 may be selected and executed.

Further, as a refinement and an extension of the method shown in fig. 1, the embodiment of the present application further provides an incoming call intention display method. Fig. 10 is a third schematic flowchart of a method for presenting an incoming call intention in an embodiment of the present application, and referring to fig. 10, the method may include:

s1001: basic information of a calling user is displayed in a first display area of an operation interface.

Step S1001 is the same as step S101 in specific implementation, and is not described here again.

S1002: and acquiring behavior data of the incoming call user according to the incoming call identification of the incoming call user.

Here, the incoming call identifier of the incoming call user may be information having a unique identity for identifying the incoming call user, such as an incoming call number, an IP address, and a user name of the incoming call user. The behavior data of the incoming call user can be history data of communication between the incoming call user and the customer service before the incoming call. For example: the voice or text content of the incoming call user which is communicated with the customer service in advance. The behavior data of the incoming call user can also be track data after the incoming call user operates in the client. For example: and the history of the pages clicked by the incoming call user in the client side is recorded. The behavior data of different incoming call users are different, so that the behavior data of the current incoming call user needs to be acquired according to the incoming call identifier of the incoming call user, and then the corresponding intention tag of the incoming call user is pushed to the customer service, so that the customer service can more efficiently serve the incoming call user.

S1003: and extracting target keywords in the behavior data.

After the behavior data of the user is acquired, there may be some useless information in the behavior data, for example: adjectives, exclamations, etc. in the voice message of the incoming user. For another example: the specific operation content (such as the specific loan total and loan year input during the calculation of monthly supply) in the trajectory data of the incoming user, and the like. Such information is not really helpful in determining the intent of the incoming call user. Therefore, it is necessary to extract words with specific practical meanings from behavior data of the calling user to form target keywords.

When extracting keywords from behavior data, different methods are required to extract keywords according to the specific form of the behavior data. For example: when the behavior data is voice information, the voice information is converted into character information, then the character information is subjected to semantic recognition, and real words are extracted to serve as target keywords. The semantic recognition can be processed by the existing semantic recognition method, such as Natural Language Processing (NLP), and will not be described herein again. For another example: when the behavior data is a history record of user operation, words conforming to a preset format can be extracted from the log and used as target keywords. Generally, the log is like a journal book, and records all operations of the user. Not all of the content in the log is helpful in determining the user's intent. The Chinese characters in the log may be search words input by the user during searching, and the codes before and after the search words do not help to determine the user intention. The operation code in the log may be followed by the content identification clicked by the user, and the operation code before the content identification is not helpful for determining the intention of the user. Therefore, keywords useful for determining the user intention can be extracted from the log according to a certain format, and the user intention can be determined.

S1004: and determining n target intention labels from the intention label library according to the similarity between the target keyword and each intention label in the intention label library.

Wherein n is a positive integer.

The intention label library is stored with various intention labels in advance, and the intention labels can maximally cover various intentions of all incoming call users. For example: consultation advance repayment intention labels, complaint business over-promotion intention labels, failure declaration intention labels, and the like.

The intent tags in the intent tag library may be obtained through a variety of channels. The intention label of the incoming call user can be manually set and added into an intention label library; the intention label of the incoming call user can also be obtained through the historical voice/character record of the incoming call user and is added into an intention label library; the intention label of the incoming call user can also be obtained through the historical operation record of the incoming call user in the client/intelligent system, and is added into an intention label library; and the intention label of the incoming call user can be acquired through the historical operation record of the customer service, and is added into an intention label library. The source of the intent tags in the intent tag library is not limited herein.

After the target keyword is obtained, firstly, calculating the similarity between the target keyword and each intention label in an intention label library; then, the n intention labels with the highest similarity are selected as the target intention labels. The target keyword may be one or more. When the target keyword is multiple, the similarity between each target keyword and each intention label in the intention label library needs to be calculated. And then selecting the n intention labels with the highest similarity as the target intention labels. The target keywords and the intention labels both belong to words, and calculating the similarity between two words is the prior art and is not described herein again.

For example, assume that intention tag library stores intention tag a, intention tag B, intention tag C. After the target keyword a and the target keyword B are obtained, the similarity between the target keyword a and the intention label A is calculated to be 90%, the similarity between the target keyword a and the intention label B is calculated to be 50%, the similarity between the target keyword a and the intention label C is calculated to be 30%, the similarity between the target keyword B and the intention label A is calculated to be 50%, the similarity between the target keyword B and the intention label B is calculated to be 80%, and the similarity between the target keyword B and the intention label C is calculated to be 20%. If 2 intent tags need to be determined, then the target intent tag is the one corresponding to the highest 90% of the similarity and the one corresponding to the second highest 80% of the similarity.

S1005: and displaying the n target intention labels in a second display area of the operation interface.

Since the n target intention labels are determined based on the behavior data of the incoming call user, the idea of the incoming call user can be represented to the greatest extent, and therefore after the n target intention labels are determined, the n intention labels are displayed to the customer service, the customer service can determine the intention of the incoming call user more quickly and accurately, and further high-quality service is provided for the incoming call user.

Further, as a refinement and extension of the method shown in fig. 10, before building the intention tag library, data required for building the intention tag library needs to be acquired in advance. Fig. 11 is a schematic flowchart of a method for acquiring behavior data of an incoming call user in an embodiment of the present application, and as shown in fig. 11, the method may include:

s1101: behavior data of all incoming call users is collected through various channels.

And the behavior data of all the incoming call users are authorized by corresponding users.

In particular, the various channels may include, but are not limited to: the system comprises a client, a customer service operating system, an online customer service system and an artificial intelligence system.

After the behavior data of the whole user is collected, the collected behavior data is not processed immediately, but is processed only when the preset time or the collected data amount reaches a certain degree, and the behavior data is collected in real time. Therefore, after some behavior data are collected, the collected behavior data are stored in a data warehouse for subsequent processing of the collected behavior data.

Fig. 12 is a schematic diagram of a data warehouse storage behavior data architecture in an embodiment of the present application, and as shown in fig. 12, after behavior data of all incoming users are collected through a client 1201, a customer service operating system 1202, an online customer service system 1203, and an artificial intelligence system 1204, the behavior data of all incoming users are stored in a data warehouse 1205, so that behavior data can be called from the data warehouse later, and an intention tag is generated.

Step S1101 may include:

s1101 a: and collecting behavior data of all incoming call users through the client.

That is, information of all user operations in the front end such as APP and H5 is collected. Specifically, the method mainly obtains the buried point data of the operation class. And the presetting is finished before the client is on line. For example: consultation payment in advance, consultation payment modes and the like. After the data of the buried point is obtained, the message is pushed through Kafka. Kafka is an existing distributed publish, subscribe messaging system. In the pushed message, there may be included: user ID, operation time, operation content, and the like. Before the message is pushed to the data warehouse, the message needs to be formatted, and the formatted message is convenient to manage. For example: the formatted message is: { "unique": 7f75973f7ae5445683baf5940f60f51a ', "mobile": 13123456789', "createTime": 2021-05-0113: 11:12 ', "operateContent": advanced payment' }.

S1101 b: and collecting behavior data of all incoming call users through the customer service operating system.

That is, in the background of the customer service operating system, data of the customer service operating for the incoming call user is collected by Binlog. Binlog is a binary log of the existing MySQL database and is used for recording SQL statement (except data query statement) information of the database operated by the user. Generally, when there is a demand from the incoming user, the incoming user may find a help solution by dialing a hotline. And the customer service can carry out corresponding operation in the customer service system to solve the problem for the calling user. Therefore, the behavior data of the incoming call user can be collected through the customer service operating system. For example: when the calling user buys a certain product in the APP, the user needs to cancel the purchase, but the user cannot cancel the purchase successfully. At the moment, the caller can contact the customer service by dialing the hotline, and the customer service can operate in the customer service operating system to help the caller to cancel purchasing the product. The customer service cancels the purchase operation in the customer service operating system and records the purchase operation in the log so as to facilitate the acquisition of the subsequent user behavior data. After behavior data of the incoming call user is acquired through the customer service operating system, the behavior data also needs to be formatted and then stored in the data warehouse.

S1101 c: and collecting behavior data of all incoming call users through the online customer service system.

That is, in the background of the online customer service system, chat data of the online customer service and the incoming call user and data of the online customer service operating for the incoming call user are collected through Binlog. Generally, when the calling user has a demand, the calling user may also find an online customer service help solution through an online service. And the on-line customer service will solve the question fed back by the user of the incoming call and perform corresponding operation in the on-line customer service system. Therefore, the behavior data of the incoming call user can be collected through the online customer service operating system. After behavior data of the incoming call user is acquired through the online customer service operating system, the behavior data also needs to be formatted and then stored in a data warehouse.

S1101 d: and collecting behavior data of all incoming call users through an artificial intelligence system.

That is, the smart voice ivr information is collected by the MirrorMaker. MirrorMaker is an existing component offered by Kafka officially for synchronization across a room. ivr, existing mutual aid voice response. Sometimes, because more caller users are consulted and fewer manual customer services are provided, an artificial intelligence system is adopted to solve the problems commonly existing in the caller users. The incoming call user can select the question to be consulted from the options given by the system, and then the artificial intelligence system gives a response to the question. The intelligent voice ivr information in the artificial intelligence system can be collected by the MirrorMaker, and the behavior data of the calling user can be obtained. After behavior data of the incoming call user is acquired through the artificial intelligence system, the behavior data also needs to be formatted and then stored in the data warehouse.

S1102: the behavioral data is formatted.

S1103: and storing the formatted behavior data to a data warehouse.

Therefore, the collection of behavior data of all incoming call users is completed, and data preparation work is well done for building the intention labels in the intention label library in the next step.

It should be noted that all the above-mentioned incoming call users refer to all the incoming call users who have used the corresponding client and contacted the customer service.

The behavior data of the incoming call user is continuously generated, and the data warehouse continuously stores the behavior data. When the residual capacity of the data warehouse is smaller than the preset capacity, the storage space of the data warehouse reaches the warning level, and in order to enable the data warehouse to continuously store the newly added behavior data, a channel with the highest priority is selected from the client, the customer service operating system, the online customer service system and the artificial intelligence system to collect the behavior data of all the incoming call users. That is, the behavior data of all the incoming users obtained from various channels is abandoned, and the behavior data of all the users is obtained by selecting individual channels, so that the coverage of the behavior data of the incoming users can be ensured, and the coverage of the intention label can be ensured.

Further, as a refinement and extension of the method shown in fig. 10, before determining the intention of the incoming call user, an intention tag library needs to be established in advance. Fig. 13 is a schematic flowchart of a method for creating an intention tag library in an embodiment of the present application, and as shown in fig. 13, the method may include:

s1301: and extracting keywords in the behavior data of all the incoming call users.

In the data warehouse, behavior data of all incoming call users are stored. And extracting all keywords from the behavior data so as to facilitate the subsequent production of the intention label.

S1302: and storing the keywords with the appearance frequency higher than the preset frequency in an intention label library as intention labels.

The behavior data contains various keywords, some keywords are not used by the calling users in practice, and may be only occasionally referred to by one or two calling users, and the corresponding performance is that the frequency of occurrence in all the extracted keywords is low. In practice, most of the incoming call users will be involved with some keywords, and the corresponding expression is that the occurrence frequency of all the extracted keywords is high. In order to enable the established intention label library to be suitable for most incoming call users and not too large (shortening subsequent matching time), all extracted keywords need to be sorted according to occurrence frequency, and keywords with occurrence frequency higher than preset frequency are selected as intention labels. The preset frequency may be determined according to actual conditions, and is not specifically limited herein.

Of course, it may also be a human adding an intent tag to the library of intent tags.

S1303: and constructing a matching rule.

Since the keywords in the behavior data of the calling user need to be matched with the intention labels in the intention label library subsequently when the intention label of the calling user is determined, after the intention label library is built, a matching rule needs to be built so as to find out the intention label of the calling user from the intention label library.

In a specific implementation process, the matching rules can be constructed according to actual situations. For example: matching the whole word of the keyword with each intention label in an intention label library; matching a first word in the keyword with each intention label in an intention label library; when the similarity between the key word and one intention label in the intention label library is 50%, the matching is considered to be successful; a match is considered successful when the keyword has 90% similarity to an intent tag in the library of intent tags. The specific content of the matching rule is not specifically limited here.

Of course, the matching rule may also be to configure a weight for each intention tag. The weight of each intent tag may be positively correlated with the frequency of occurrence of the intent tag in the user behavior data.

S1304: the matching rules are stored in an intent tag library.

It should be noted that steps S1301 to S1302 and steps S1303 to S1304 may be performed synchronously or asynchronously, and are not limited herein.

The above is the process of building the intent tag library. After the establishment of the intention label library is completed, the intention labels in the intention label library are required to be updated subsequently.

Whenever the incoming call user generates new behavior data, the newly generated behavior data is stored in the data warehouse. Specifically, the time may be preset, for example: and acquiring newly-added behavior data in the data warehouse every 10 nights, extracting keywords in the behavior data, and storing the keywords with the occurrence frequency higher than the preset frequency as the intention labels to be added in a specified position so that business personnel can check whether the keywords can be added into the intention label library.

For the voice data of the caller and the customer service stored in the data warehouse, the voice data can be converted into text data, keywords are extracted from the text data, and then the intention label to be added is determined and stored in the designated position.

Here, the high frequency of occurrence of a keyword of the same content means a plurality of keywords. When the keyword is used as an intention label, duplication removal is required, and only one keyword is required to be reserved.

Besides updating the intention labels in the intention label library, the weights of the intention labels in the intention label library can be reconfigured according to actual conditions, and therefore the accuracy of the recommended intention labels is improved.

Specifically, the weight may be configured for each intention label in the intention label library according to an intention type with the largest proportion among the actual intentions of the incoming call user within a preset time period. Wherein the more similar the intent tag type in the intent tag library, the higher the weight corresponding to the intent tag. That is, the real intentions of the incoming call user of the incoming call in the latest event are collected and classified to obtain various intention types (such as complaint type, consultation type, etc.). And in the intention label library, the weight of the intention label corresponding to the intention type with the largest occupation ratio is improved. For example: and increasing the weight of the intention label of the complaint class in the intention label library when the complaint of the incoming call user is more recently.

At this point, the updating process of the intention label in the intention label library is also completed.

Further, as a refinement and extension of the method shown in fig. 10, the embodiment of the present application also provides an incoming call intention determination method. Fig. 14 is a fourth schematic flowchart of a method for displaying an incoming call intention in an embodiment of the present application, and referring to fig. 14, the method may include:

s1401: basic information of a calling user is displayed in a first display area of an operation interface.

Step S1401 is the same as the specific implementation of step S101, and is not described herein again.

S1402: and when the incoming call user is connected with the intelligent voice, acquiring behavior data of the incoming call user according to the incoming call identification of the incoming call user.

When the calling user has an appeal and needs to find a communication solution of the calling user, the calling user may dial a customer service hotline (95 XXX). After dialing the customer service hotline, the user of the incoming call will generally first present a voice prompt. When the caller user listens to the voice prompt, the behavior data of the caller is obtained according to the caller identification of the caller, so that the determined intention label of the caller can be immediately displayed to the customer service when the customer service actually establishes a call with the caller. The method and the system avoid that the intention label of the incoming call user is determined only after the customer service actually establishes a call with the incoming call user, and improve the speed of displaying the intention label to the customer service.

In practical application, the latest 20 pieces of behavior data of the incoming call user can be acquired.

The specific manner of acquiring the behavior data of the incoming call user is the same as step S1002, and is not described herein again.

S1403: and extracting target keywords in the behavior data.

Specifically, step S1403 may include:

s1403 a: all keywords in the behavior data are extracted.

And extracting how many keywords can be extracted from the behavior data, and taking the keywords as all extracted keywords.

S1403 b: and deleting the keywords which do not appear in the first keyword library in all the keywords.

The first keyword library stores a large number of words in a dictionary.

Since the behavior data of the incoming call user includes the language of the incoming call user, and the language word is too spoken, or includes a large number of adjectives unrelated to the determination of the intention tag, it is necessary to delete all keywords that do not appear in the first keyword library, so as to improve the accuracy of the keywords, and further facilitate the determination of the intention tag.

Of course, the first keyword library may also store related words in the domain to which the customer service belongs. Therefore, words irrelevant to the industry can be filtered, and the matching efficiency of the intention label is improved. The specific content stored in the first keyword library is not limited herein.

S1403 c: and adding the keywords with similarity higher than a preset value with all the keywords in the second keyword library into all the keywords.

And the second keyword library stores the similar meaning words of various words.

And adding the keywords with similarity higher than a preset value with all the keywords in the second keyword library into all the keywords, so that the success rate of matching the intention labels can be improved, and the accuracy of intention determination is further improved.

Of course, the second keyword library may also store word pairs with corresponding relationships. The corresponding relation can be artificially matched according to the actual service scene. For example: the incoming user often mentions the word a too spoken, while in the art the word b is often used. Then, the word a is associated with the word b. And when the word a exists in all the keywords, adding the word b into all the keywords according to the corresponding relation.

Words in the second keyword library, which have corresponding relations with all the keywords, are added to all the keywords, so that the success rate of matching the intention labels can be improved, and the accuracy of intention determination is further improved.

S1403 d: and taking all the deleted and/or added keywords as target keywords.

It should be noted that, the step S1403b and the step S1403c may be executed at the same time or alternatively, and are not limited specifically here.

S1404: and calculating the similarity between the target keyword and each intention label in the intention label library.

And there is a corresponding weight for each intention tag in the library of intention tags.

S1405: and determining n target intention labels from the intention label library according to the product of the similarity and the corresponding weight.

Wherein the n target intention labels are the n intention labels with the maximum product value in the intention label library.

For example, assume that the target keyword is keyword a, the intention tag library stores intention tag a, corresponding weight is 20, intention tag B, corresponding weight is 50, intention tag C, corresponding weight is 80. Through calculation, the similarity between the keyword a and the intention label A is 30%, the similarity between the keyword a and the intention label B is 60%, and the similarity between the keyword a and the intention label C is 90%. The similarity is multiplied by the corresponding weight again, and the result of keyword a and intention label a is 30% × 20 ═ 6, the result of keyword a and intention label B is 60% × 50 ═ 30, and the result of keyword a and intention label C is 90% × 80 ═ 48. If 1 target intent tag is determined, it is intent tag C.

In practical application, 7 intention labels can be recommended according to practical situations for customer service reference.

When the number of the target keywords is plural, if the top-ranked intention labels are recommended according to the similarity and the weight, there may be a case where the recommended n intention labels are related to only some keywords of the target keywords, and other keywords are not taken into consideration, and the keywords that are not taken into consideration may be the intention of the current incoming call user. In view of this, in selecting intent tags from the library of intent tags, one or more top-ranked intent tags may be selected from the library of intent tags for each of the target keywords.

S1406: and displaying the n target intention labels in a second display area of the operation interface.

After n target intention labels are determined, the n target intention labels can be displayed in a certain area of the customer service interface according to the high-low order of the scores of the target intention labels.

S1407: and obtaining a feedback result of the customer service.

After the customer service finishes communicating with the incoming call user, the customer service can determine whether the n recommended target intention labels are correct according to the actual communication condition with the incoming call user, and then the n recommended target intention labels are used as feedback results to feed back.

S1408: when the feedback result indicates that the n target intention labels are inaccurate, adding a new intention label to an intention label library; or, the weight corresponding to each intention label in the intention label library is reconfigured.

When the feedback result indicates that the n target intention labels are inaccurate, in order to enable the intention labels to be recommended to the customer service accurately subsequently, the intention labels in the intention label library can be adjusted.

The first adjustment mode is as follows: adding a new intent tag to the library of intent tags. And the new intent tag is related to the feedback result. That is, when the customer service finds that the intention label recommended to the customer service is incorrect, a new intention label can be written and fed back according to the actual result of communication with the incoming call user. And subsequently adding the new intention label fed back by the incoming call user into the intention label library according to actual conditions.

The second adjustment mode is as follows: and reconfiguring the corresponding weight of each intention label in the intention label library. The intention tag recommendation is inaccurate, and it is possible that the matching rule (here, the configured weight for the intention tag) is not suitable, so the corresponding weight of each intention tag in the intention tag library can also be reconfigured, so that the intention tag can be accurately recommended to the customer service in the following. The specific weight of the configuration can be configured according to the actual experience of the administrator.

Finally, the architecture of the incoming call intention display method provided by the embodiment of the present application is described, so as to more clearly understand the execution process of the incoming call intention display method provided by the embodiment of the present application.

Fig. 15 is a schematic structural diagram of the incoming call intention display method in the embodiment of the present application, and referring to fig. 14, the incoming call intention display method provided in the embodiment of the present application is mainly divided into four layers, that is, an operation interaction layer, a service layer, an algorithm layer, and a storage layer. In the storage layer, a database is configured for storing relevant data. The database may include: an intent tag repository and a data repository. In the algorithm layer, various algorithms are configured for intention tag creation, rule generation, and incoming call intention generation, respectively. In the business layer, an interface is configured to specify the incoming call intention matching priority (i.e., weight). In the operation interaction layer, each functional module is configured. The functions of each functional module comprise: incoming call intention display, intention tag confirmation, incoming call intention matching rule maintenance and the like.

Based on the same inventive concept, as an implementation of the method, the embodiment of the application also provides a device for displaying the incoming call intention. Fig. 16 is a first structural schematic diagram of the incoming call intention display device in the embodiment of the present application, and referring to fig. 16, the device may include:

the first display module 1601 is configured to display basic information of a calling user in a first display area of an operation interface, where the basic information is obtained by matching a calling identifier of the calling user.

A second display module 1602, configured to display an intention tag of the incoming call user in a second display area of the operation interface before or when the incoming call is successfully connected, where the intention tag is used to represent the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

Further, as a refinement and extension of the device shown in fig. 16, the embodiment of the present application also provides a device for presenting an incoming call intention. Fig. 17 is a schematic structural diagram of a second incoming call intention display device in the embodiment of the present application, and referring to fig. 17, the device may include:

the first display module 1701 is configured to display basic information of a calling user in a first display area of an operation interface, where the basic information is obtained by matching a calling identifier of the calling user.

A second display module 1702, configured to display an intention tag of the incoming call user in a second display area of the operation interface before or when the incoming call is successfully connected, where the intention tag is used to represent the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

A third displaying module 1703, configured to display the dialect corresponding to the intention tag in a third display area of the operation interface, where the dialect of the third display area corresponds to the intention tag of the second display area.

A first adjusting module 1704, configured to adjust a ranking of each telephone operation corresponding to the intention tag, where the ranking of each telephone operation is related to an actual telephone operation of the customer service, and the actual telephone operation is obtained through current communication content of the customer service and the incoming call user.

A second adjusting module 1705, configured to adjust the ranking of each intention tag in the second display area, or highlight a part of the intention tags in the second display area, where the ranking of each intention tag and the highlighted part of the intention tags are related to an actual purpose of the incoming call user, and the actual purpose is obtained through current communication content of the incoming call user.

A fourth displaying module 1706, configured to display the phrase in the current communication content of the incoming call user in a fourth display area of the operation interface.

A fifth displaying module 1707, configured to display, in a fifth display area of an operation interface, a word in the word group that does not appear in the intention label of the second display area, where an editing window is disposed in the fifth display area, and a customer service can edit the word in the fifth display area through the editing window to generate a new intention label.

A sixth displaying module 1708, configured to display, in a sixth display area of the operation interface, a professional term corresponding to a word in the phrase that does not appear in the intention tag of the second display area.

Further, as a refinement and extension of the device shown in fig. 16, the embodiment of the present application also provides a device for presenting an incoming call intention. Fig. 18 is a schematic structural diagram of a third incoming call intention display device in an embodiment of the present application, and referring to fig. 18, the device may include:

the library building module 1801 includes:

the collecting unit 1801a is configured to collect behavior data of all incoming users through multiple channels.

The collecting unit 1801a is specifically configured to, when the remaining capacity of the data warehouse for storing behavior data is smaller than a preset capacity, select a channel with the highest priority from the client, the customer service operating system, the online customer service system, and the artificial intelligence system to collect behavior data of all incoming call users.

The second extracting unit 1801b is configured to extract keywords in the behavior data of all incoming users.

A storage unit 1801c, configured to store the keyword with the occurrence frequency higher than the preset frequency as an intention tag in an intention tag library.

A configuration module 1802, configured to configure a weight for each intention label in the intention label library according to an intention type with the largest proportion among actual intentions of the incoming call user within a preset time period, where the weight corresponding to an intention label with a more similar intention type in the intention label library is higher.

The first displaying module 1803 is configured to display basic information of the incoming call user in a first display area of the operation interface, where the basic information is obtained by matching an incoming call identifier of the incoming call user.

The obtaining module 1804 is specifically configured to obtain behavior data of the incoming call user according to the identification of the incoming call when the incoming call user is connected with the intelligent voice.

The extraction module 1805 includes:

a first extracting unit 1805a, configured to extract all keywords in the behavior data.

An increasing/decreasing unit 1805b is configured to delete a keyword that does not appear in the first keyword library from all the keywords, and/or add a keyword having a similarity higher than a preset value to all the keywords from the second keyword library to all the keywords.

A determining unit 1805c, configured to use all deleted and/or added keywords as target keywords.

The matching module 1806 includes:

a calculating unit 1806a, configured to calculate similarity between the target keyword and each intention label in the intention label library.

A matching unit 1806b, configured to determine n target intention labels from the intention label library according to a product of the similarity and the corresponding weight, where the n target intention labels are the n intention labels with the largest product value in the intention label library.

A second display module 1807, configured to display an intention tag of the incoming call user in a second display area of the operation interface before or when the incoming call is successfully connected, where the intention tag is used to represent a purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

The adjusting module 1808 includes:

the obtaining unit 1808a is configured to obtain a feedback result of the customer service.

An adjusting unit 1808b, configured to add a new intention tag to the intention tag library when the feedback result indicates that the n target intention tags are inaccurate, where the new intention tag is related to the feedback result; or, the weight corresponding to each intention label in the intention label library is reconfigured.

It is to be noted here that the above description of the embodiments of the apparatus, similar to the description of the embodiments of the method described above, has similar advantageous effects as the embodiments of the method. For technical details not disclosed in the embodiments of the apparatus of the present application, reference is made to the description of the embodiments of the method of the present application for understanding.

Based on the same inventive concept, the embodiment of the application also provides the electronic equipment. Fig. 19 is a schematic structural diagram of an electronic device in an embodiment of the present application, and referring to fig. 19, the electronic device may include: a processor 1901, memory 1902, bus 1903; the processor 1901 and the memory 1902 complete communication with each other through the bus 1903; the processor 1901 is configured to call program instructions in the memory 1902 to perform methods in one or more of the embodiments described above.

It is to be noted here that the above description of the embodiments of the electronic device, similar to the description of the embodiments of the method described above, has similar advantageous effects as the embodiments of the method. For technical details not disclosed in the embodiments of the electronic device of the present application, refer to the description of the embodiments of the method of the present application for understanding.

Based on the same inventive concept, the embodiment of the present application further provides a computer-readable storage medium, where the storage medium may include: a stored program; wherein the program controls the device on which the storage medium is located to execute the method in one or more of the above embodiments when the program runs.

It is to be noted here that the above description of the storage medium embodiments, like the description of the above method embodiments, has similar advantageous effects as the method embodiments. For technical details not disclosed in the embodiments of the storage medium of the present application, reference is made to the description of the embodiments of the method of the present application for understanding.

The above description is only for the specific embodiments of the present application, but the scope of the present application is not limited thereto, and any person skilled in the art can easily conceive of the changes or substitutions within the technical scope of the present application, and shall be covered by the scope of the present application. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.

Claims

1. A method for displaying an incoming call intention is characterized by comprising the following steps:

basic information of a calling user is displayed in a first display area of an operation interface, and the basic information is obtained through matching of a calling identification of the calling user;

before or when the incoming call is connected to be successful, displaying an intention label of the incoming call user in a second display area of the operation interface, wherein the intention label is used for representing the purpose of the incoming call user;

the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

2. The method of claim 1, wherein before or when the incoming call is successfully connected, the method further comprises:

and displaying the dialect corresponding to the intention label in a third display area of an operation interface, wherein the dialect of the third display area corresponds to the intention label of the second display area.

3. The method of claim 2, wherein after the presenting the words corresponding to the intent tags in a third display area of an operator interface, the method further comprises:

and adjusting the sequence of each telephone operation corresponding to the intention label, wherein the sequence of each telephone operation is related to the actual telephone operation of the customer service, and the actual telephone operation is obtained through the current communication content of the customer service and the incoming call user.

4. The method of claim 1, wherein after the intent tag of the incoming call user is presented in the second display area of the operator interface, the method further comprises:

and adjusting the sequence of each intention label in the second display area, or highlighting part of intention labels in the second display area, wherein the sequence of each intention label and the highlighted part of intention labels are related to the actual purpose of the incoming call user, and the actual purpose is obtained through the current communication content of the incoming call user.

5. The method of claim 1, wherein after the incoming electrical connection is successful, the method further comprises:

displaying the phrase in the current communication content of the incoming call user in a fourth display area of the operation interface; or,

displaying words in the intention labels, which do not appear in the second display area, in the phrase in a fifth display area of an operation interface, wherein an editing window is arranged in the fifth display area, and a customer service can edit the words in the fifth display area through the editing window to generate new intention labels; or,

and displaying the professional terms corresponding to the words which do not appear in the intention labels of the second display area in the phrases in a sixth display area of an operation interface.

6. The method according to claim 1, wherein the presenting the intention tag of the incoming call user in a second display area of the operation interface comprises:

acquiring behavior data of the incoming call user according to the incoming call identification;

extracting target keywords in the behavior data;

determining n target intention labels from the intention label library according to the similarity between the target keyword and each intention label in the intention label library, wherein n is a positive integer;

and displaying the n target intention labels in a second display area of the operation interface.

7. The method of claim 6, wherein there is a corresponding weight for each of the intent tags in the library of intent tags; determining n target intention labels from the intention label library according to the similarity between the target keyword and each intention label in the intention label library, wherein the method comprises the following steps:

calculating the similarity between the target keyword and each intention label in an intention label library;

and determining the n target intention labels from the intention label library according to the product of the similarity and the corresponding weight, wherein the n target intention labels are the n intention labels with the maximum product value in the intention label library.

8. The method of claim 6, wherein the extracting the target keyword from the behavior data comprises:

extracting all keywords in the behavior data;

deleting the keywords which do not appear in the first keyword library in all the keywords, and/or adding the keywords with similarity higher than a preset value with all the keywords in the second keyword library into all the keywords;

and taking all the deleted and/or added keywords as the target keywords.

9. The method according to claim 6, wherein the obtaining the behavior data of the incoming call user according to the incoming call identifier comprises:

and when the incoming call user is connected with the intelligent voice, acquiring the behavior data of the incoming call user according to the incoming call identification.

10. The method according to any one of claims 6 to 9, wherein before said determining n target intention labels from the library of intention labels according to the similarity of the target keyword to each intention label in the library of intention labels, the method further comprises:

collecting behavior data of all incoming call users through various channels, wherein the behavior data of all the incoming call users are authorized by corresponding users;

extracting key words in the behavior data of all the incoming call users;

and storing the keywords with the appearance frequency higher than the preset frequency in the intention label library as intention labels.

11. The method of claim 10, wherein the plurality of channels comprises at least: the system comprises a client, a customer service operating system, an online customer service system and an artificial intelligence system; the collecting of behavior data of all incoming call users through multiple channels comprises the following steps:

and when the residual capacity of the data warehouse for storing the behavior data is smaller than the preset capacity, selecting a channel with the highest priority from the client, the customer service operating system, the online customer service system and the artificial intelligence system to collect the behavior data of all the incoming call users.

12. The method according to claim 10, wherein after storing the keywords having the occurrence frequency higher than the preset frequency as the intention tags in the intention tag library, the method further comprises:

configuring a weight for each intention label in the intention label library according to an intention type with the largest proportion in the actual intentions of the incoming call user in a preset time period, wherein the more similar intention labels to the intention type in the intention label library, the higher the corresponding weight is.

13. The method according to any one of claims 6 to 9, wherein after said displaying said n target intent tags, said method further comprises:

obtaining a feedback result of the customer service;

when the feedback result indicates that the n target intention labels are inaccurate, adding a new intention label to the intention label library, wherein the new intention label is related to the feedback result.

14. The method according to any one of claims 7 to 9, wherein after said displaying said n target intent tags, said method further comprises:

obtaining a feedback result of the customer service;

when the feedback result indicates that the n target intention labels are inaccurate, reconfiguring the weight corresponding to each intention label in the intention label library.

15. An incoming call intention display device, the device comprising:

the first display module is used for displaying basic information of the incoming call user in a first display area of an operation interface, and the basic information is obtained by matching incoming call identification of the incoming call user;

the second display module is used for displaying an intention label of the incoming call user in a second display area of the operation interface before the incoming call is connected to be successful or when the incoming call is connected to be successful, wherein the intention label is used for representing the purpose of the incoming call user; the intention label is obtained by extracting behavior data of the incoming call user, and the behavior data is obtained based on the basic information or the incoming call identification matching.

16. An electronic device, comprising: a processor, a memory, a bus;

the processor and the memory complete mutual communication through the bus; the processor is configured to invoke program instructions in the memory to perform the method of any of claims 1 to 14.

17. A computer-readable storage medium, comprising: a stored program; wherein the program, when executed, controls the device on which the storage medium is located to perform the method of any one of claims 1 to 14.