A method and device for intelligent review invitation

By inputting the first customer service conversation data into the trained invitation and evaluation judgment model, we determine whether invitation and evaluation information needs to be sent, and the satisfaction distortion caused by the subjective invitation and evaluation of the customer service is solved, and a true reflection of the customer service level is achieved.

CN112866492BActive Publication Date: 2024-12-06BEIJING WODONG TIANJUN INFORMATION TECH CO LTD +1
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN201911173285.8
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2019-11-26
Publication Date
2024-12-06
Estimated Expiration
2039-11-26

AI Technical Summary

Technical Problem

In the prior art, the subjective invitation of customer service to comments leads to distortion of user satisfaction and cannot truly reflect the level of customer service.

Method used

By inputting the first customer service conversation data into the invitation judgment model obtained through the second customer service conversation data training, it is determined whether the invitation comment information needs to be sent, and the system automatically intelligent invitation comment is realized to reduce the influence of subjective factors.

Benefits of technology

It realizes a true reflection of customer service satisfaction, reduces satisfaction distortion caused by customer service subjective invitations, and improves the authenticity of service level.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN112866492B_ABST
    Figure CN112866492B_ABST
Patent Text Reader

Abstract

The present invention discloses a method and device for intelligently inviting reviews, and relates to the field of computer technology. A specific implementation of the method includes: inputting first customer service conversation data into an invitation and review judgment model to determine whether it is necessary to send an invitation and review message to the customer corresponding to the first customer service conversation, wherein the invitation and review judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is the customer service conversation data generated online in real time, and the second customer service conversation data is the customer service conversation data in the most recent time period; if necessary, the invitation and review message is sent to the terminal corresponding to the customer. This implementation realizes automatic and intelligent invitation and review by the system, and reduces the situation where the customer service satisfaction is distorted due to the subjective invitation and review by the customer service, and the service level of the customer service cannot be truly reflected.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The present invention relates to the field of computer technology, and in particular to a method and device for intelligently inviting comments. Background Art

[0002] In the manual dialogue system, when the customer service has helped the customer to solve the customer's consultation problem and confirmed that the conversation can be ended, the customer service will proactively send an invitation for evaluation so that the customer can evaluate the customer service's service, and the evaluation result (i.e. user satisfaction) will be used as a business indicator for evaluating customer service services.

[0003] In the process of implementing the present invention, the inventors found that there are at least the following problems in the prior art:

[0004] The existing review invitation method in which customer service sends a review invitation selectively invites reviews for the conversation based on the customer's consultation and service situation. This may lead to a deviation from the actual user satisfaction and cannot truly reflect the user's satisfaction with the customer service. Summary of the invention

[0005] In view of this, an embodiment of the present invention provides a method and device for intelligent review invitation, which can realize automatic intelligent review invitation by the system, reduce the situation where customer service satisfaction is distorted due to subjective review invitation by customer service, and cannot truly reflect the service level of customer service.

[0006] To achieve the above objective, according to one aspect of an embodiment of the present invention, a method for intelligently inviting reviews is provided.

[0007] A method for intelligently inviting reviews, comprising: inputting first customer service conversation data into an invitation review judgment model to judge whether it is necessary to send an invitation review message to a customer corresponding to the first customer service conversation, wherein the invitation review judgment model is obtained by training second customer service conversation data, the first customer service conversation data is customer service conversation data generated online in real time, and the second customer service conversation data is customer service conversation data within a recent time period; if necessary, sending the invitation review message to a terminal corresponding to the customer.

[0008] Optionally, training the second customer service conversation data includes: obtaining the second customer service conversation data within a recent time period; processing the second customer service conversation data according to preset dimensional indicators to obtain sample data, and the preset dimensional indicators include data integrity, data validity and whether the data complies with business process specifications; and obtaining an invitation to review judgment model by training the sample data.

[0009] Optionally, processing the second customer service conversation data according to preset dimensional indicators to obtain sample data includes: judging for each conversation data in the second customer service conversation data whether all of the following conditions are met: the conversation length corresponding to the conversation data reaches a set first length threshold, the conversation data does not include invalid consulting questions and answers, and the logic of the conversation data complies with business standard service process specifications; if all of the conditions are met, judging whether the conversation length corresponding to the conversation data exceeds a set second length threshold; if not, taking the conversation data as sample data; otherwise, intercepting the conversation data to obtain conversation data that meets the first length threshold, and taking the intercepted conversation data as sample data.

[0010] Optionally, the customer service conversation data includes a session identifier of the customer service conversation; and after sending the review invitation information to the terminal corresponding to the customer, it also includes: obtaining the session identifier of the first customer service conversation through data embedding; obtaining the first customer service conversation data and the review invitation judgment result corresponding to the first customer service conversation data from the customer service conversation log according to the session identifier; and adjusting the parameters of the review invitation judgment model by analyzing the first customer service conversation data and the review invitation judgment result.

[0011] Optionally, analyzing the first customer service conversation data and the result of the review invitation judgment to adjust the parameters of the review invitation judgment model includes: judging whether the first customer service conversation data meets the requirements of the sample data according to preset dimensional indicators; if so, judging whether the result of the review invitation judgment of the first customer service conversation data is "review invitation"; if so, taking the first customer service conversation data and the result of the review invitation judgment as positive samples, otherwise, taking the first customer service conversation data and the result of the review invitation judgment as negative samples; using the sample data corresponding to the first customer service conversation data to adjust the parameters of the review invitation judgment model.

[0012] Optionally, after sending the review invitation information to the terminal corresponding to the customer, it also includes: obtaining the customer's evaluation index value through data embedding, and analyzing the customer service satisfaction of the customer service according to the evaluation index value.

[0013] According to another aspect of an embodiment of the present invention, a device for intelligently inviting comments is provided.

[0014] A device for intelligently inviting reviews, comprising: an invitation review judgment module, used to input first customer service conversation data into an invitation review judgment model to judge whether it is necessary to send an invitation review message to a customer corresponding to the first customer service conversation, wherein the invitation review judgment model is obtained by training second customer service conversation data, the first customer service conversation data is customer service conversation data generated online in real time, and the second customer service conversation data is customer service conversation data within a recent time period; an invitation review sending module, used to send the invitation review message to a terminal corresponding to the customer if necessary.

[0015] Optionally, it also includes a model training module for training the review invitation judgment model, including: obtaining second customer service conversation data within a recent time period; processing the second customer service conversation data according to preset dimensional indicators to obtain sample data, and the preset dimensional indicators include data integrity, data validity and whether the data complies with business process specifications; and obtaining the review invitation judgment model by training the sample data.

[0016] Optionally, the model training module is also used to: determine for each conversation data in the second customer service conversation data whether all of the following conditions are met: the conversation length corresponding to the conversation data reaches a set first length threshold, the conversation data does not include invalid consulting questions and answers, and the logic of the conversation data complies with business standard service process specifications; if all of the conditions are met, determine whether the conversation length corresponding to the conversation data exceeds a set second length threshold; if not, use the conversation data as sample data; otherwise, intercept the conversation data to obtain conversation data that meets the first length threshold, and use the intercepted conversation data as sample data.

[0017] Optionally, the customer service conversation data includes a session identifier of the customer service conversation; and the device also includes a model adjustment module, which is used to: after sending the review invitation information to the terminal corresponding to the customer, obtain the session identifier of the first customer service conversation through data embedding; obtain the first customer service conversation data and the review invitation judgment result corresponding to the first customer service conversation data from the customer service conversation log according to the session identifier; and adjust the parameters of the review invitation judgment model by analyzing the first customer service conversation data and the review invitation judgment result.

[0018] Optionally, the model adjustment module is also used to: determine whether the first customer service conversation data meets the requirements of the sample data according to preset dimensional indicators; if so, determine whether the invitation and review judgment result of the first customer service conversation data is "invitation and review"; if so, use the first customer service conversation data and its invitation and review judgment result as positive samples, otherwise, use the first customer service conversation data and its invitation and review judgment result as negative samples; use the sample data corresponding to the first customer service conversation data to adjust the parameters of the invitation and review judgment model.

[0019] Optionally, it also includes an evaluation acquisition module, which is used to: after sending the evaluation invitation information to the terminal corresponding to the customer, obtain the customer's evaluation index value through data embedding, and analyze the customer service satisfaction based on the evaluation index value.

[0020] According to another aspect of an embodiment of the present invention, an electronic device for intelligently inviting comments is provided.

[0021] An electronic device for intelligent review invitation includes: one or more processors; a storage device for storing one or more programs, when the one or more programs are executed by the one or more processors, the one or more processors implement the intelligent review invitation method provided by an embodiment of the present invention.

[0022] According to still another aspect of the embodiments of the present invention, a computer readable medium is provided.

[0023] A computer-readable medium stores a computer program, which, when executed by a processor, implements the intelligent review invitation method provided in an embodiment of the present invention.

[0024] An embodiment of the above invention has the following advantages or beneficial effects: by inputting the first customer service conversation data generated online in real time into the invitation review judgment model to judge whether it is necessary to send the invitation review information to the customer corresponding to the first customer service conversation, wherein the invitation review judgment model is obtained by training the second customer service conversation data in the most recent time period; if necessary, the invitation review information is sent to the terminal corresponding to the customer, so as to judge whether it is necessary to send the invitation review information through the invitation review judgment model, realize the automatic intelligent invitation review by the system, and reduce the distortion of customer service satisfaction due to the subjective invitation review of the customer service, which cannot truly reflect the service level of the customer service.

[0025] The further effects of the above-mentioned non-conventional optional manner will be described below in conjunction with the specific implementation manner. BRIEF DESCRIPTION OF THE DRAWINGS

[0026] The accompanying drawings are used to better understand the present invention and do not constitute an improper limitation of the present invention.

[0027] Figure 1 is a schematic diagram of the main steps of the method for intelligent review invitation according to an embodiment of the present invention;

[0028] Figure 2 It is a schematic diagram of the standard service process of customer service business;

[0029] Figure 3 is a schematic diagram of the training process of the review invitation judgment model according to an embodiment of the present invention;

[0030] Figure 4 1 is a schematic diagram of the implementation flow of the method for intelligent review invitation according to an embodiment of the present invention;

[0031] Figure 5 is a schematic diagram of main modules of a device for intelligent review invitation according to an embodiment of the present invention;

[0032] Figure 6 is an exemplary system architecture diagram to which embodiments of the present invention may be applied;

[0033] Figure 7 It is a schematic diagram of the structure of a computer system of a terminal device or a server suitable for implementing an embodiment of the present invention. DETAILED DESCRIPTION

[0034] The following is a description of exemplary embodiments of the present invention in conjunction with the accompanying drawings, including various details of the embodiments of the present invention to facilitate understanding, which should be considered as merely exemplary. Therefore, it should be recognized by those of ordinary skill in the art that various changes and modifications may be made to the embodiments described herein without departing from the scope and spirit of the present invention. Similarly, for clarity and conciseness, the description of well-known functions and structures is omitted in the following description.

[0035] Figure 1 FIG. 1 is a schematic diagram of the main steps of the method for intelligent review invitation according to an embodiment of the present invention. Figure 1 As shown, the intelligent review invitation method of the embodiment of the present invention mainly includes the following steps S101 and S102.

[0036] Step S101: inputting the first customer service conversation data into the review invitation judgment model to judge whether it is necessary to send a review invitation message to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is the customer service conversation data generated online in real time, and the second customer service conversation data is the customer service conversation data in the most recent time period;

[0037] Step S102: If necessary, the review invitation information is sent to the terminal corresponding to the client.

[0038] According to the above-mentioned steps S101 and S102, the system can automatically and intelligently invite reviews, thereby reducing the situation where customer service satisfaction is distorted due to subjective customer service invitations, and the situation where the customer service level cannot be truly reflected.

[0039] According to an embodiment of the present invention, when training the review invitation judgment model, training the second customer service dialogue data may specifically include:

[0040] Obtain the second customer service conversation data within the most recent time period;

[0041] The second customer service conversation data is processed according to preset dimension indicators to obtain sample data, where the preset dimension indicators include data integrity, data validity, and whether the data complies with business process specifications;

[0042] The invitation review judgment model is obtained by training the sample data.

[0043] In the above-mentioned embodiment, when obtaining the second customer service conversation data, historical conversation data within the last week or the last month can be extracted, for example, the second customer service conversation data can be obtained through the conversation log, specifically including: the time when the customer service conversation occurred, the session identification ID corresponding to the customer service conversation, the customer identity information, the customer consultation question, and the customer service reply and other data information. Afterwards, the obtained second customer service conversation data is judged, screened, intercepted, etc. according to the preset dimensional indicators of the present invention, so as to obtain sample data. For example: for users who do not have any consultation questions after coming in, do not explain the consultation intention after the consultation, or do not confirm whether the reply given by the customer service has solved the problem, etc., these customer service conversation data need to be deleted and cannot be used as sample data.

[0044] According to one embodiment of the present invention, the review invitation judgment model is trained by using the Hierarchical Attention Network (HAN) network structure algorithm. The features of the HAN model are: the hierarchical layers use word-level layers and sentence-level layers, which conform to the document structure; and the attention mechanism is used to intuitively see the importance of each sentence and word when constructing a text representation using this model. The attention mechanism can be added to the importance of words and sentences, so that the customer's intention can be better judged. In addition, other model frameworks can also be used to implement it, such as classification algorithm models, etc.

[0045] In an embodiment of the present invention, processing the second customer service dialogue data according to a preset dimension indicator to obtain sample data may specifically include:

[0046] For each dialogue data in the second customer service dialogue data, it is determined whether all of the following conditions are met: the session length corresponding to the dialogue data reaches a set first length threshold, the dialogue data does not include invalid consulting questions and answers, and the logic of the dialogue data complies with the business standard service process specification;

[0047] If all the conditions are met, determining whether the session length corresponding to the conversation data exceeds a set second length threshold;

[0048] If it does not exceed, the conversation data is used as sample data;

[0049] Otherwise, the conversation data is intercepted to obtain conversation data that meets the first length threshold, and the intercepted conversation data is used as sample data.

[0050] In this embodiment, when processing the second customer service conversation data, generally, if the conversation length corresponding to the conversation data is too short, it will be impossible to accurately judge the customer service situation. Therefore, it is necessary to select conversation data whose conversation length reaches the set first length threshold, where the conversation length refers to the sum of the number of messages sent between the customer and the customer service; the first length threshold is, for example, 8 messages, which can be flexibly adjusted according to the needs of actual applications. Since invalid consulting questions and replies cannot reflect the customer service situation, it is necessary to delete this part of the conversation data. For example, messages sent by the system, chat data between customers and customer service, etc. are all invalid consulting questions and replies. In addition, conversation data that does not meet the business standard service process specifications cannot be used as training samples. Figure 2 This is a diagram of the standard service process for customer service. Figure 2 As shown, when a customer comes in (i.e., enters the customer service consultation page), the customer service needs to first confirm the customer's intention, then provide a solution, and ask the customer whether he accepts it; if the customer accepts it, the customer service will be asked again whether he has any other questions to consult, until the customer has no questions to consult; if the customer does not accept it, other solutions will be provided again, and the user will be asked whether he accepts it.

[0051] Figure 3 FIG. 1 is a schematic diagram of the training process of the review invitation judgment model according to an embodiment of the present invention. Figure 3As shown, in an embodiment of the present invention, by extracting the customer service conversation log within a recent period of time, and then removing the system messages and meaningless messages in the log, the continuous customer or customer service messages are merged. Specifically, the messages can be merged into an object for program processing according to the time sequence of the messages sent by the customer and the customer service. The continuous here refers to continuous in time sequence. Afterwards, it is determined whether the session length of each customer service conversation data reaches a specified first threshold (for example: 10); if not, the customer service conversation data is deleted, otherwise, it is further determined whether the session length exceeds a second threshold (for example: 12). If not, the customer service conversation data is used as a training sample. If it exceeds, the customer service conversation data is intercepted to obtain customer service conversation data that meets the first threshold length, and the intercepted customer service conversation data is used as a training sample. When intercepting the customer service conversation data, in order to better express the customer's intention characteristics, part of the data can be intercepted forward and part of the data can be intercepted reversely. Forward interception refers to intercepting from front to back according to the time sequence of the messages sent by the customer and the customer service, and reverse interception refers to intercepting from back to front according to the time sequence of the messages sent by the customer and the customer service. Generally, if 10 messages need to be intercepted, 7 messages can be intercepted in the forward direction and 3 messages can be intercepted in the reverse direction. The specific number of intercepted messages can be flexibly set according to needs.

[0052] After the training samples are generated, the training samples can be used to train the review invitation judgment model. After the model is trained, the customer service conversation data within the specified time period is called through the interface to determine whether it is necessary to invite reviews, and then manually mark whether the review result is correct. Usually, the negative sample data with incorrect judgments in the output results of the model judgment is analyzed, and the model is continuously optimized to be as optimal as possible while meeting the business objectives. After that, it is determined whether the review invitation accuracy rate reaches the optimal value (or whether it reaches the set target value). If so, the model is used as the trained review invitation judgment model. Otherwise, the algorithm parameters are adjusted until the evaluation accuracy rate reaches the optimal value (or reaches the set target value).

[0053] In another embodiment of the present invention, the customer service conversation data includes a session identifier of the customer service conversation; and after sending the review invitation information to the terminal corresponding to the customer in step S102, the following operations may also be included:

[0054] Obtaining a session identifier of the first customer service conversation through data tracking;

[0055] Acquire the first customer service conversation data and the evaluation invitation judgment result corresponding to the first customer service conversation data from the customer service conversation log according to the conversation identifier;

[0056] The parameters of the review invitation judgment model are adjusted by analyzing the first customer service dialogue data and the review invitation judgment result.

[0057] In the above embodiment, when analyzing the first customer service dialogue data and the review invitation judgment result to adjust the parameters of the review invitation judgment model, the following steps may be specifically included:

[0058] Determine whether the first customer service conversation data meets the sample data requirement according to the preset dimension indicators;

[0059] If satisfied, determining whether the evaluation invitation determination result of the first customer service conversation data is "evaluation invitation";

[0060] If yes, the first customer service conversation data and the result of the evaluation invitation are used as positive samples; otherwise, the first customer service conversation data and the result of the evaluation invitation are used as negative samples;

[0061] The sample data corresponding to the first customer service conversation data is used to adjust the parameters of the review invitation judgment model.

[0062] In the above-mentioned embodiment, the parameters of the review invitation judgment model are adjusted according to the online first customer service conversation data and its review invitation judgment result, thereby realizing continuous optimization of the model according to the real-time data generated online, so that the trained model can better adapt to the online customer service review invitation.

[0063] According to the technical solution of the present invention, after the review invitation information is sent to the terminal corresponding to the customer, the customer's evaluation index value can also be obtained through data embedding, and the customer service satisfaction is analyzed based on the evaluation index value. After the user makes an evaluation, the embedding data is extracted from the big data platform for statistical analysis of the overall customer service satisfaction and the service satisfaction of a single customer service. For example, a customer service representative has 30 automatic review invitations for a conversation in a day. The service satisfaction of a single customer service representative on that day is the sum of the satisfaction of the 30 conversations divided by 30. The calculation method of the overall customer service satisfaction is similar.

[0064] Figure 4 FIG. 1 is a schematic diagram of the implementation flow of the intelligent review invitation method according to an embodiment of the present invention. Figure 4 As shown, when the conversation begins, the customer asks questions and the customer service replies, and then determines whether the customer responds. If not, the process ends without inviting a review. If so, the review invitation judgment model determines whether the customer service conversation data meets the conditions for sending a review invitation message. If not, the customer continues to ask questions and the customer service replies. If so, a review invitation message is automatically sent. Then it determines whether the user has commented. If the user has commented, the evaluation index value is obtained. Otherwise, the evaluation index value fails to be obtained and the process ends.

[0065] Figure 5Schematic diagram of the main modules of the apparatus for intelligent review invitation according to an embodiment of the present invention. Figure 5 As shown, the intelligent review invitation device 500 of the embodiment of the present invention mainly includes a review invitation judgment module 501 and a review invitation sending module 502.

[0066] The review invitation judgment module 501 is used to input the first customer service conversation data into the review invitation judgment model to judge whether it is necessary to send a review invitation message to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is the customer service conversation data generated online in real time, and the second customer service conversation data is the customer service conversation data in the most recent time period;

[0067] The review invitation sending module 502 is used to send the review invitation information to the terminal corresponding to the client if necessary.

[0068] According to an embodiment of the present invention, the intelligent review invitation device 500 may further include a model training module (not shown in the figure) for training the review invitation judgment model, specifically including:

[0069] Obtain the second customer service conversation data within the most recent time period;

[0070] The second customer service conversation data is processed according to preset dimension indicators to obtain sample data, wherein the preset dimension indicators include data integrity, data validity, and whether the data complies with business process specifications;

[0071] The review invitation judgment model is obtained by training the sample data.

[0072] According to another embodiment of the present invention, the model training module may also be used for:

[0073] For each dialogue data in the second customer service dialogue data, determining whether all of the following conditions are met: the session length corresponding to the dialogue data reaches a set first length threshold, the dialogue data does not include invalid consulting questions and answers, and the logic of the dialogue data complies with business standard service process specifications;

[0074] If all conditions are met, determining whether the session length corresponding to the conversation data exceeds a set second length threshold;

[0075] If not, the conversation data is used as sample data;

[0076] Otherwise, the conversation data is intercepted to obtain conversation data that meets the first length threshold, and the intercepted conversation data is used as sample data.

[0077] According to another embodiment of the present invention, the customer service dialogue data includes a session identifier of the customer service dialogue; and the intelligent review invitation device 500 may also include a model adjustment module (not shown in the figure) for:

[0078] After sending the review invitation information to the terminal corresponding to the customer, obtaining the session identifier of the first customer service dialogue through data embedding;

[0079] Acquire the first customer service conversation data and the evaluation invitation judgment result corresponding to the first customer service conversation data from the customer service conversation log according to the conversation identifier;

[0080] The parameters of the review invitation judgment model are adjusted by analyzing the first customer service dialogue data and the review invitation judgment result.

[0081] According to yet another embodiment of the present invention, the model adjustment module may also be used for:

[0082] Determining whether the first customer service conversation data meets the sample data requirement according to a preset dimension indicator;

[0083] If satisfied, determining whether the evaluation invitation determination result of the first customer service conversation data is "evaluation invitation";

[0084] If yes, the first customer service conversation data and the result of the evaluation invitation are used as positive samples; otherwise, the first customer service conversation data and the result of the evaluation invitation are used as negative samples;

[0085] The sample data corresponding to the first customer service dialogue data is used to adjust parameters of the review invitation judgment model.

[0086] According to the technical solution of the present invention, the intelligent evaluation invitation device 500 may further include an evaluation acquisition module (not shown in the figure) for:

[0087] After the review invitation information is sent to the terminal corresponding to the customer, the customer's evaluation index value is obtained through data embedding, and the customer service satisfaction is analyzed based on the evaluation index value.

[0088] According to the technical solution of the embodiment of the present invention, the first customer service conversation data generated online in real time is input into the review invitation judgment model to determine whether it is necessary to send review invitation information to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data in the most recent time period; if necessary, the review invitation information is sent to the terminal corresponding to the customer, thereby realizing the use of the review invitation judgment model to determine whether it is necessary to send the review invitation information, realizing automatic and intelligent review invitation by the system, and reducing the situation where customer service satisfaction is distorted due to subjective review invitation by customer service, and the service level of the customer service cannot be truly reflected.

[0089] Figure 6 An exemplary system architecture 600 is shown to which the method for intelligently inviting reviews or the apparatus for intelligently inviting reviews according to an embodiment of the present invention may be applied.

[0090] like Figure 6 As shown, system architecture 600 may include terminal devices 601, 602, 603, network 604 and server 605. Network 604 is used to provide a medium for communication links between terminal devices 601, 602, 603 and server 605. Network 604 may include various connection types, such as wired, wireless communication links or optical fiber cables, etc.

[0091] Users can use terminal devices 601, 602, and 603 to interact with server 605 through network 604 to receive or send messages, etc. Various communication client applications can be installed on terminal devices 601, 602, and 603, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (only examples).

[0092] The terminal devices 601 , 602 , and 603 may be various electronic devices having a display screen and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, and desktop computers, etc.

[0093] The server 605 may be a server that provides various services, such as a backend management server (only an example) that provides support for shopping websites browsed by users using the terminal devices 601, 602, and 603. The backend management server may analyze and process the received data such as product information query requests, and feed back the processing results (such as target push information, product information - only an example) to the terminal device.

[0094] It should be noted that the method for intelligently inviting reviews provided in the embodiment of the present invention is generally executed by the server 605 , and accordingly, the device for intelligently inviting reviews is generally set in the server 605 .

[0095] It should be understood that Figure 6The number of terminal devices, networks and servers in the embodiment is only for illustration. Any number of terminal devices, networks and servers may be provided according to implementation requirements.

[0096] Reference below Figure 7 , which shows a schematic diagram of the structure of a computer system 700 of a terminal device or a server suitable for implementing an embodiment of the present invention. Figure 7 The terminal device or server shown is only an example and should not bring any limitation to the functions and scope of use of the embodiments of the present invention.

[0097] like Figure 7 As shown, the computer system 700 includes a central processing unit (CPU) 701, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 702 or a program loaded from a storage part 708 into a random access memory (RAM) 703. In the RAM 703, various programs and data required for the operation of the system 700 are also stored. The CPU 701, the ROM 702, and the RAM 703 are connected to each other via a bus 704. An input / output (I / O) interface 705 is also connected to the bus 704.

[0098] The following components are connected to the I / O interface 705: an input section 706 including a keyboard, a mouse, etc.; an output section 707 including a cathode ray tube (CRT), a liquid crystal display (LCD), etc., and a speaker, etc.; a storage section 708 including a hard disk, etc.; and a communication section 709 including a network interface card such as a LAN card, a modem, etc. The communication section 709 performs communication processing via a network such as the Internet. A drive 710 is also connected to the I / O interface 705 as needed. A removable medium 711, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 710 as needed, so that a computer program read therefrom is installed into the storage section 708 as needed.

[0099] In particular, according to the embodiments disclosed in the present invention, the process described above with reference to the flowchart can be implemented as a computer software program. For example, the embodiments disclosed in the present invention include a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program includes a program code for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication part 709, and / or installed from the removable medium 711. When the computer program is executed by the central processing unit (CPU) 701, the above-mentioned functions defined in the system of the present invention are executed.

[0100] It should be noted that the computer-readable medium shown in the present invention may be a computer-readable signal medium or a computer-readable storage medium or any combination of the above two. The computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, device or device, or any combination of the above. More specific examples of computer-readable storage media may include, but are not limited to: an electrical connection with one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), an optical fiber, a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in combination with an instruction execution system, device or device. In the present invention, a computer-readable signal medium may include a data signal propagated in a baseband or as part of a carrier wave, which carries a computer-readable program code. This propagated data signal may take a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination of the above. Computer-readable signal media may also be any computer-readable medium other than computer-readable storage media, which may send, propagate or transmit a program for use by or in conjunction with an instruction execution system, apparatus or device. The program code contained on the computer-readable medium may be transmitted using any appropriate medium, including but not limited to: wireless, wire, optical cable, RF, etc., or any suitable combination of the above.

[0101] The flow chart and block diagram in the accompanying drawings illustrate the possible architecture, function and operation of the system, method and computer program product according to various embodiments of the present invention. In this regard, each box in the flow chart or block diagram can represent a module, a program segment, or a part of a code, and the above-mentioned module, program segment, or a part of a code contains one or more executable instructions for realizing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in a different order from the order marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flow chart, and the combination of the boxes in the block diagram or flow chart can be implemented with a dedicated hardware-based system that performs a specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0102] The units or modules involved in the embodiments of the present invention may be implemented in software or hardware. The units or modules described may also be provided in a processor. For example, they may be described as follows: a processor includes a review invitation judgment module and a review invitation sending module. The names of these units or modules do not, in some cases, constitute limitations on the units or modules themselves. For example, the review invitation judgment module may also be described as "a module for inputting the first customer service conversation data into the review invitation judgment model to determine whether it is necessary to send review invitation information to the customer corresponding to the first customer service conversation."

[0103] As another aspect, the present invention also provides a computer-readable medium, which may be included in the device described in the above embodiment; or it may exist independently without being assembled into the device. The above computer-readable medium carries one or more programs, and when the above one or more programs are executed by a device, the device includes: inputting the first customer service conversation data into the review invitation judgment model to judge whether it is necessary to send a review invitation message to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is the customer service conversation data generated online in real time, and the second customer service conversation data is the customer service conversation data in the most recent time period; if necessary, the review invitation message is sent to the terminal corresponding to the customer.

[0104] According to the technical solution of the embodiment of the present invention, the first customer service conversation data generated online in real time is input into the review invitation judgment model to determine whether it is necessary to send review invitation information to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data in the most recent time period; if necessary, the review invitation information is sent to the terminal corresponding to the customer, thereby realizing the use of the review invitation judgment model to determine whether it is necessary to send the review invitation information, realizing automatic and intelligent review invitation by the system, and reducing the situation where customer service satisfaction is distorted due to subjective review invitation by customer service, and the service level of the customer service cannot be truly reflected.

[0105] The above specific implementations do not constitute a limitation on the protection scope of the present invention. It should be understood by those skilled in the art that various modifications, combinations, sub-combinations and substitutions may occur depending on design requirements and other factors. Any modification, equivalent substitution and improvement made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A method for intelligently inviting comments. It is characterized in that include: Inputting the first customer service conversation data into a review invitation judgment model to judge whether it is necessary to send a review invitation message to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is customer service conversation data generated online in real time, and the second customer service conversation data is customer service conversation data in a recent time period; If necessary, the review invitation information is sent to the terminal corresponding to the client; The training of the second customer service conversation data includes: obtaining the second customer service conversation data in a recent time period; processing the second customer service conversation data according to preset dimension indicators to obtain sample data, wherein the preset dimension indicators include data integrity, data validity, and whether the data complies with business process specifications; The review invitation judgment model is obtained by training the sample data.

2. The method according to claim 1, It is characterized in that Processing the second customer service conversation data according to a preset dimension indicator to obtain sample data includes: For each conversation data in the second customer service conversation data, determine whether all of the following conditions are met: the conversation length corresponding to the conversation data reaches a set first length threshold, the conversation data does not include invalid consulting questions and answers, and the logic of the conversation data complies with business standard service process specifications; if all of the conditions are met, determine whether the conversation length corresponding to the conversation data exceeds a set second length threshold; if not, use the conversation data as sample data; otherwise, intercept the conversation data to obtain conversation data that meets the first length threshold, and use the intercepted conversation data as sample data.

3. The method according to claim 1, It is characterized in that The customer service conversation data includes a session identifier of the customer service conversation; Furthermore, after sending the review invitation information to the terminal corresponding to the client, the method further includes: Obtaining a session identifier of the first customer service conversation through data embedding; Acquire the first customer service conversation data and the evaluation invitation judgment result corresponding to the first customer service conversation data from the customer service conversation log according to the conversation identifier; The parameters of the review invitation judgment model are adjusted by analyzing the first customer service dialogue data and the review invitation judgment result.

4. The method according to claim 3, It is characterized in that Analyzing the first customer service dialogue data and the review invitation judgment result thereof to adjust the parameters of the review invitation judgment model includes: Determining whether the first customer service conversation data meets the sample data requirement according to a preset dimension indicator; If satisfied, determining whether the evaluation invitation determination result of the first customer service conversation data is "evaluation invitation"; If yes, the first customer service conversation data and the result of the evaluation invitation are used as positive samples; otherwise, the first customer service conversation data and the result of the evaluation invitation are used as negative samples; The sample data corresponding to the first customer service dialogue data is used to adjust parameters of the review invitation judgment model.

5. The method according to claim 1, It is characterized in that After sending the review invitation information to the terminal corresponding to the client, the method further includes: The customer's evaluation index values ​​are obtained through data embedding, and the customer service satisfaction is analyzed based on the evaluation index values.

6. A device for intelligently inviting comments. It is characterized in that include: a review invitation judgment module, used for inputting the first customer service conversation data into the review invitation judgment model to judge whether it is necessary to send a review invitation message to the customer corresponding to the first customer service conversation, wherein the review invitation judgment model is obtained by training the second customer service conversation data, the first customer service conversation data is the customer service conversation data generated online in real time, and the second customer service conversation data is the customer service conversation data in a recent time period; A review invitation sending module, used to send the review invitation information to the terminal corresponding to the client if necessary; The model training module is used to train the review invitation judgment model, including: obtaining the second customer service conversation data in the most recent time period; processing the second customer service conversation data according to preset dimensional indicators to obtain sample data, and the preset dimensional indicators include data integrity, data validity and whether the data complies with business process specifications; and obtaining the review invitation judgment model by training the sample data.

7. An electronic device for intelligent review invitation, It is characterized in that include: one or more processors; a storage device for storing one or more programs, When the one or more programs are executed by the one or more processors, the one or more processors implement the method according to any one of claims 1 to 5.

8. A computer readable medium having a computer program stored thereon, It is characterized in that When the program is executed by a processor, the method according to any one of claims 1 to 5 is implemented.