CN114036379A - Customer service recommendation method and device, electronic equipment and readable storage medium - Google Patents
Customer service recommendation method and device, electronic equipment and readable storage medium Download PDFInfo
- Publication number
- CN114036379A CN114036379A CN202111306314.0A CN202111306314A CN114036379A CN 114036379 A CN114036379 A CN 114036379A CN 202111306314 A CN202111306314 A CN 202111306314A CN 114036379 A CN114036379 A CN 114036379A
- Authority
- CN
- China
- Prior art keywords
- consultation
- customer service
- target
- data
- historical
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 54
- 238000013145 classification model Methods 0.000 claims abstract description 38
- 238000012549 training Methods 0.000 claims description 35
- 230000000977 initiatory effect Effects 0.000 claims description 28
- 238000005516 engineering process Methods 0.000 claims description 17
- 230000008569 process Effects 0.000 claims description 13
- 238000004590 computer program Methods 0.000 claims description 9
- 238000012015 optical character recognition Methods 0.000 claims description 8
- 230000004044 response Effects 0.000 claims description 6
- 238000009223 counseling Methods 0.000 description 6
- 230000009471 action Effects 0.000 description 3
- 238000012545 processing Methods 0.000 description 3
- 230000002159 abnormal effect Effects 0.000 description 2
- 230000006854 communication Effects 0.000 description 2
- 235000014510 cooky Nutrition 0.000 description 2
- 238000010586 diagram Methods 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000006243 chemical reaction Methods 0.000 description 1
- 239000003795 chemical substances by application Substances 0.000 description 1
- 238000004140 cleaning Methods 0.000 description 1
- 230000003749 cleanliness Effects 0.000 description 1
- 238000013135 deep learning Methods 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000000605 extraction Methods 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000002372 labelling Methods 0.000 description 1
- 238000005065 mining Methods 0.000 description 1
- 230000003287 optical effect Effects 0.000 description 1
- 230000000750 progressive effect Effects 0.000 description 1
- 230000011218 segmentation Effects 0.000 description 1
- 239000013598 vector Substances 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/90—Details of database functions independent of the retrieved data types
- G06F16/95—Retrieval from the web
- G06F16/953—Querying, e.g. by the use of web search engines
- G06F16/9535—Search customisation based on user profiles and personalisation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q30/00—Commerce
- G06Q30/01—Customer relationship services
- G06Q30/015—Providing customer assistance, e.g. assisting a customer within a business location or via helpdesk
- G06Q30/016—After-sales
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Databases & Information Systems (AREA)
- General Physics & Mathematics (AREA)
- Business, Economics & Management (AREA)
- General Engineering & Computer Science (AREA)
- Bioinformatics & Computational Biology (AREA)
- Development Economics (AREA)
- Evolutionary Biology (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Artificial Intelligence (AREA)
- Accounting & Taxation (AREA)
- Evolutionary Computation (AREA)
- Economics (AREA)
- Finance (AREA)
- Marketing (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The application discloses a customer service recommendation method, a customer service recommendation device, electronic equipment and a readable storage medium, wherein the customer service recommendation method comprises the following steps: acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request; inputting historical consultation data into a preset consultation problem classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories; receiving a target problem category output by the consultation problem classification model; and placing the customer service consultation request into a target customer service consultation queue corresponding to the target problem category. The method can efficiently and accurately identify the target problem category of the customer service consultation initiated by the user at this time without the need of actively inputting any selection information by the user, so that the user can queue in the customer service queues of corresponding categories, the probability of transferring orders and queuing again is reduced, and the customer service consultation experience of the user is improved.
Description
Technical Field
The present application relates to the field of data classification technologies, and in particular, to a method and an apparatus for recommending customer service, an electronic device, and a computer-readable storage medium.
Background
In a customer service system (such as proprietary customer service in the finance and tax field) with a high specialty in the consultation problem, if a user needs manual service, the system often classifies the consultation user into a queue of random customer service in a random order classification mode, and the consultation user can only be passively served by the manual customer service in a queue mode. Therefore, the consulting customer service cannot know the field of the problem to be consulted, and the user cannot know the good field of the consulting customer service, so that the conditions of wasting time and customer service resources such as transferring orders and queuing again are likely to occur, the utilization of the consulting customer service resources is not facilitated, and the consulting experience of the user is also influenced.
It should be noted that, because the consulted problem relates to a strong specialty, the stronger the specialty, the more closely the problem categories which look like the actual difference is large are subdivided, and these problem categories are difficult to be selected by the user through the category labels composed of a few characters, because the user who initiates the consultation to the customer service is also difficult to classify the problem which the user wants to consult according to these category labels, the problem of transferring order and queuing again still exists.
Therefore, how to provide a better customer service recommendation scheme for a customer service system with a stronger specialty for consultation problems is a problem to be solved urgently by technical personnel in the field.
Disclosure of Invention
The application aims to provide a customer service recommendation method and device, electronic equipment and a computer-readable storage medium.
In a first aspect, the present application provides a customer service recommendation method, including:
acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
inputting historical consultation data into a preset consultation problem classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories;
receiving a target problem category output by the consultation problem classification model;
and placing the customer service consultation request into a target customer service consultation queue corresponding to the target problem category.
Optionally, obtaining historical consultation data from a historical consultation session with customer service according to the received customer service consultation request includes:
determining a consultation initiating object according to the received customer service consultation request;
extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
and summarizing all original consultation data in the form of texts to obtain historical consultation data.
Optionally, the method further includes:
in response to the absence of a historical consultation session corresponding to the consultation initiating object, one of the alternative plurality of question categories is determined as a target question category.
Optionally, determining one of the multiple candidate problem categories as a target problem category includes:
determining a highest-heat problem category corresponding to the current time period from a plurality of problem categories which are selected;
and determining the highest-heat problem category as the target problem category.
Optionally, after the customer service consultation request is placed in the target customer service consultation queue corresponding to the target problem category, the method further includes:
controlling the target customer service corresponding to the target customer service consultation queue to process each customer service consultation request in the queue in sequence;
the control target customer service marks the target problem category on the consultation content corresponding to the target problem category in the current consultation data after consultation is finished, and a new training sample is obtained;
the control target customer service marks the non-target problem category of the consultation content which does not correspond to the target problem category in the current consultation data after consultation is finished, and a new negative training sample is obtained;
and continuously iterating and updating the consultation problem classification model by using the new positive training sample and the new negative training sample.
The present application provides, in a second aspect, a customer service recommendation device, including:
the historical consultation data acquisition unit is configured for acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
the consultation problem classification calling model is configured for inputting historical consultation data into a preset consultation problem classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories;
a target problem category receiving unit configured to receive a target problem category output by the consultation problem classification model;
and the targeted queuing unit is configured to place the customer service consultation request into a target customer service consultation queue corresponding to the target problem category.
Optionally, the historical consulting data obtaining unit is further configured to:
determining a consultation initiating object according to the received customer service consultation request;
extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
and summarizing all original consultation data in the form of texts to obtain historical consultation data.
Optionally, the apparatus further comprises:
and the first consultation target problem category determining unit is configured to determine one of the plurality of alternative problem categories as the target problem category in response to the absence of the historical consultation session corresponding to the consultation initiating object.
Optionally, the first consultation target problem category determining unit is further configured to:
determining a highest-heat problem category corresponding to the current time period from a plurality of problem categories which are selected;
and determining the highest-heat problem category as the target problem category.
Optionally, the apparatus further comprises:
after the customer service consultation requests are placed in a target customer service consultation queue corresponding to the target problem category, controlling target customer service treatment corresponding to the target customer service consultation queue to sequentially process each customer service consultation request in the queue;
the control target customer service marks the target problem category on the consultation content corresponding to the target problem category in the current consultation data after consultation is finished, and a new training sample is obtained;
the control target customer service marks the non-target problem category of the consultation content which does not correspond to the target problem category in the current consultation data after consultation is finished, and a new negative training sample is obtained;
and continuously iterating and updating the consultation problem classification model by using the new positive training sample and the new negative training sample.
In a third aspect, the present application provides an electronic device comprising:
a memory for a computer program;
a processor for implementing the steps of the customer service recommendation method as described above when executing the computer program.
In a fourth aspect, a computer-readable storage medium is provided, on which a computer program is stored, which, when being executed by a processor, carries out the steps of the customer service recommendation method as described above.
According to the customer service recommendation method, aiming at customer service scenes with high professional consulting problems, by collecting consulting data of the same user history and customer service and by means of a preset problem consulting category model capable of representing corresponding relations between different historical consulting data and the problem categories, the target problem category of the customer service consultation initiated by the user at this time can be efficiently and accurately identified without the need of the user to actively input any selection information, so that the user can queue in the corresponding category of customer service queues, the probability of transferring orders and queuing again is reduced, and the customer service consultation experience of the user is improved. The application also provides a deployment system, a deployment device and a computer-readable storage medium of the service in the cloud environment, which have the beneficial effects and are not repeated herein.
Drawings
In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings needed to be used in the description of the embodiments or the prior art will be briefly introduced below, it is obvious that the drawings in the following description are only embodiments of the present application, and for those skilled in the art, other drawings can be obtained according to the provided drawings without creative efforts.
Fig. 1 is a flowchart of a customer service recommendation method according to an embodiment of the present application;
fig. 2 is a flowchart of a method for obtaining historical consulting data in a customer service recommendation method according to an embodiment of the present application;
fig. 3 is a flowchart of a method for continuously iterating and updating a consultation problem classification model in a customer service recommendation method provided in an embodiment of the present application;
fig. 4 is a block diagram of a structure of a customer service recommendation device according to an embodiment of the present application.
Detailed Description
The core of the application is to provide a method, a system and a device for deploying services in a cloud environment and a computer-readable storage medium.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Referring to fig. 1, fig. 1 is a flowchart of a customer service recommendation method according to an embodiment of the present application, where the process 100 includes the following steps:
step 101: acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
this step is intended to acquire, by an executing agent (e.g., a server, a terminal device providing a customer service to an accessing user) adapted to execute the customer service recommendation method, previous consultation data of the user from a previous historical consultation session with the customer service of the user in response to the received customer service consultation request.
It should be noted that, because the present application is directed to a consultation problem related to a customer service scenario with a strong specialty, the customer service consultation request does not include guidance information that can be used to extract information to help determine which type of customer service should be selected for consultation, and the customer service consultation request only simply corresponds to a user triggering the customer service consultation service, for example, clicking a "consult to customer service" function button on a webpage provided by an executive subject to the outside.
The historical consultation session may be determined based on user identity information (e.g., Cookie information recorded by a browser, logged-in registered identity information, etc.) included in the customer service consultation request, and the historical consultation data may be recorded at the user side or on the storage unit of the execution main body.
Step 102: inputting historical consultation data into a preset consultation problem classification model;
on the basis of step 101, this step is intended to input historical consultation data into a preset consultation problem classification model by the above-mentioned executive body. The consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the affiliated problem categories.
Step 103: receiving a target problem category output by the consultation problem classification model;
the consultation problem classification model is a pre-training class model constructed based on a deep learning thought, and is characterized in that training and learning are continuously performed in a training sample which is hidden with corresponding relations between different historical consultation data and the classes of the problems, so that the finally obtained consultation problem classification model has certain capacity of judging which class of problems the actually input consultation data correspond to.
Generally, the expression form of historical consulting data includes texts, images, voices and the like, the consulting problem classification model can convert the texts, images and voices which are convenient for a user to recognize through a built-in feature processing layer into a form which is convenient for a computer to recognize, such as vectors, matrixes, lattices and the like, then a special feature extraction layer is used for deeply mining the association of features of different layers and dimensions on a labeling result (namely, which consulting data correspond to which problem categories), and further a universal corresponding relation is learned from a training sample, so that the consulting problem classification model has the capability of outputting a more accurate classification result by a new sample which is not included in the training sample.
It should be noted that, if there is no historical consultation session corresponding to the consultation initiating object (i.e., the user initiates a customer service consultation request for the first time), one of the multiple candidate question categories is determined as the target question category. Specifically, how to select or select the question category from the multiple candidate question categories may be multiple, for example, the question category is selected according to the heat degree of the current time period, for example, the highest heat question category corresponding to the current time period (i.e., the question category selected or determined to belong to by the most users) is determined from the multiple candidate question categories, and the highest heat question category is determined as the target question category. Of course, other principles may be used, such as the problem category corresponding to the customer service consultation queue with the least number of people currently queued, and so on.
Step 104: and placing the customer service consultation request into a target customer service consultation queue corresponding to the target problem category.
After the target problem category corresponding to the customer service consultation request is obtained, the execution main body firstly determines a target customer service consultation queue corresponding to the target problem category, and then places the customer service consultation request into the target customer service consultation queue for arrangement so as to wait for being processed through the first-in first-out characteristic of the queue.
That is, different customer services are configured for different problem categories in advance, and the number of the customer services configured for one problem category may also be multiple (for example, determined according to the frequency, the duty ratio, and the like of the consultation of the problem category), when one problem category corresponds to multiple customer services, the current customer service consultation request may be placed in the queue of the customer service with the least number of requests in the queue in combination with load balancing, or the customer service consultation request may be placed in the same customer service in the history consultation session in combination with consultation continuity, but only in the case that the current consultation request and the history consultation request both correspond to the same problem category, and the same customer service is currently in the order receiving state.
Based on the technical scheme, the customer service recommendation method provided by the embodiment of the application aims at a customer service scene with a relatively high specialty of the consultation problem, and can efficiently and accurately identify the target problem category of the customer service consultation initiated by the user at this time without actively inputting any selection information by collecting the consultation data of the same user history and customer service and by means of a preset problem consultation category model capable of representing the corresponding relation between different history consultation data and the affiliated problem category, so that the user can queue in the customer service queue of the corresponding category, the probability of transferring orders and queuing again is reduced, and the customer service consultation experience of the user is improved.
Referring to fig. 2, fig. 2 is a flowchart of a method for obtaining historical consulting data in a customer service recommendation method provided in an embodiment of the present application, where the process 200 includes the following steps:
step 201: determining a consultation initiating object according to the received customer service consultation request;
the execution subject determines a consultation initiating object for initiating the customer service consultation request according to the received customer service consultation request. Specifically, the consultation initiating object can be determined by inquiring Cookie information recorded by the browser, logged-in registered identity information and the like, or can be determined temporarily by requiring the user to input a mobile phone number, a user name and the like of the user.
It should be noted that, the determination of the consultation initiating object is to query the historical consultation session of the consultation initiating object before the consultation initiating object through the determined relevant information of the consultation initiating object so as to obtain the historical consultation data. And thus can be used to determine the consultation initiating object as long as it can be used to determine the means of historical consultation sessions.
Step 202: extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
on the basis of step 201, this step is intended to extract at least one kind of original counseling data including text, image, voice from the historical counseling session of the counseling initiation object and the customer service by the execution subject. Specifically, the original consultation data in the text form is generally a text conversation between the user and the customer service, the original consultation data in the image form is generally a screenshot representing the transaction intention or recording problems of the customer service or the user collected from various channels in the communication process, and the original consultation data in the voice form is generally a voice signal recorded by the customer service or the user in the communication process.
Specifically, the original advisory data in different expression forms are extracted in different ways, and the original advisory data in the form of text can be directly extracted from the text dialogue, while the original advisory data in the form of image can be obtained by downloading the original image or extracting the original image in the form of screenshot, and the original advisory data in the form of voice can be obtained by reading the voice signal or re-encoding and decoding the audio.
Step 203: converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
step 204: converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
in order to unify the presentation form of the original counseling data and facilitate the computer to understand the contents contained in the original counseling data, steps 203 and 204 respectively provide unified conversion into text form through an optical character recognition technology (OCR) and a voice recognition technology for the original counseling data in an image form and a voice form.
The speech recognition technology can also be embodied as a speech content recognition model constructed based on a natural speech understanding technology so as to improve the accuracy of speech recognition.
Step 205: and summarizing all original consultation data in the form of texts to obtain historical consultation data.
On the basis of the steps 203 and 204, the step aims to summarize all the original advisory data expressed in the form of text by the executive body to obtain historical advisory data.
Specifically, in the summarizing process, for convenience of subsequent processing, word segmentation processing can be performed on the original consulting data in all text forms to segment and obtain consulting words representing the consulting characteristics. Furthermore, the data can be subdivided into core words, non-core words, invalid contents, abnormal characters and the like, so that data cleaning is performed by combining subdivision types of the core words, the non-core words, the invalid contents, the abnormal characters and the like, the 'cleanliness degree' of data input into the consultation problem classification model is improved, and the accuracy of output results is further improved.
Referring to fig. 3, fig. 3 is a flowchart of a method for continuously iterating and updating a consulting problem classification model in a customer service recommendation method provided in an embodiment of the present application, where the process 300 includes the following steps:
step 301: controlling the target customer service corresponding to the target customer service consultation queue to process each customer service consultation request in the queue in sequence;
namely, the target customer service processes each customer service consultation request in the queue in turn according to the default queue characteristic of first-in first-out. In addition, if the customer service consultation request of some users has special authority, the ordering of the customer service consultation request in the queue can be adjusted according to the special authority.
Step 302: the control target customer service marks the target problem category on the consultation content corresponding to the target problem category in the current consultation data after consultation is finished, and a new training sample is obtained;
the step aims to control the target customer service to re-label the consultation content which can represent the target problem and interest categories in the current consultation data in the consultation finishing state by the executive main body (namely, the part of the consultation content is labeled as belonging to the target problem categories, and the judgment on the problem categories of the model at the beginning is correct), so that the consultation content is used as a new positive training sample of a follow-up training consultation problem classification model, and the new positive training sample contains some new features which are not learned to the corresponding target problem categories in previous model training.
Step 303: the control target customer service marks the non-target problem category of the consultation content which does not correspond to the target problem category in the current consultation data after consultation is finished, and a new negative training sample is obtained;
different from step 302, in this step, the execution subject controls the target customer service to re-label the consulting content capable of representing the consulting data not conforming to the target problem category in the current consulting data in the consulting ending state (i.e. label the consulting content as not belonging to the target problem category, and explain that the problem category of the model at first is judged incorrectly), so as to serve as a new negative training sample of the subsequent training consulting problem classification model, and in order to make the model learn a new feature that the same consulting data is not judged as the target problem category any more subsequently in the new negative training sample.
Step 304: and continuously iterating and updating the consultation problem classification model by using the new positive training sample and the new negative training sample.
On the basis of the steps 302 and 303, the step aims to continuously iterate and update the consultation problem classification model by the executive body by using a new positive training sample and a new negative training sample.
Specifically, the subsequent iteration and updating of the consulting problem classification model can be performed once every a period of time, and whether the updating requirement is met can be determined according to the number of the accumulated new training samples, and the like.
Because the situation is complicated and cannot be illustrated by a list, a person skilled in the art can realize that many examples exist according to the basic method principle provided by the application and the practical situation, and the protection scope of the application should be protected without enough inventive work.
Referring to fig. 4, fig. 4 is a block diagram illustrating a structure of a customer service recommendation device according to an embodiment of the present application, where the customer service recommendation device 400 includes the following functional units:
a historical consultation data acquisition unit 401 configured to acquire historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
a consultation question classification calling model 402 configured to input historical consultation data into a preset consultation question classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories;
a target question category receiving unit 403 configured to receive a target question category output by the consultation question classification model;
a targeted queuing unit 404 configured to place the customer service consultation request into a target customer service consultation queue corresponding to the target problem category.
Wherein, the historical consulting data obtaining unit 401 may be further configured to:
determining a consultation initiating object according to the received customer service consultation request;
extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
and summarizing all original consultation data in the form of texts to obtain historical consultation data.
Further, the customer service recommendation device 400 may further include:
and the first consultation target problem category determining unit is configured to determine one of the plurality of alternative problem categories as the target problem category in response to the absence of the historical consultation session corresponding to the consultation initiating object.
Wherein the first consultation target problem category determining unit may be further configured to:
determining a highest-heat problem category corresponding to the current time period from a plurality of problem categories which are selected;
and determining the highest-heat problem category as the target problem category.
Further, the customer service recommendation device 400 may further include:
after the customer service consultation requests are placed in a target customer service consultation queue corresponding to the target problem category, controlling target customer service treatment corresponding to the target customer service consultation queue to sequentially process each customer service consultation request in the queue;
the control target customer service marks the target problem category on the consultation content corresponding to the target problem category in the current consultation data after consultation is finished, and a new training sample is obtained;
the control target customer service marks the non-target problem category of the consultation content which does not correspond to the target problem category in the current consultation data after consultation is finished, and a new negative training sample is obtained;
and continuously iterating and updating the consultation problem classification model by using the new positive training sample and the new negative training sample.
The present embodiment is an embodiment of an apparatus corresponding to the above-described embodiment of the method.
According to the customer service recommendation device provided by the embodiment of the application, aiming at a customer service scene with a relatively high specialty of consultation problems, by collecting consultation data of the same user history and customer service and by means of a preset problem consultation category model capable of representing the corresponding relation between different history consultation data and the corresponding problem categories, the target problem category of the user initiating the customer service consultation at this time can be identified efficiently and accurately without requiring the user to actively input any selection information, so that the user can queue in the corresponding category of customer service queues, the probability of order change and queuing again is reduced, and the customer service consultation experience of the user is improved.
Based on the foregoing embodiment, the present application further provides a deployment apparatus for services in a cloud environment, where the deployment apparatus may include a memory and a processor, where the memory stores a computer program, and when the processor calls the computer program in the memory, the steps provided in the foregoing embodiment may be implemented. Of course, the deployment device may also include various necessary network interfaces, power supplies, and other components.
The present application also provides a computer-readable storage medium, on which a computer program is stored, which, when executed by an execution terminal or processor, can implement the steps provided by the above-mentioned embodiments. The storage medium may include: various media capable of storing program codes, such as a usb disk, a removable hard disk, a Read-Only Memory (ROM), a Random Access Memory (RAM), a magnetic disk, or an optical disk.
The embodiments are described in a progressive manner in the specification, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. The device disclosed by the embodiment corresponds to the method disclosed by the embodiment, so that the description is simple, and the relevant points can be referred to the method part for description.
Those of skill would further appreciate that the various illustrative elements and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the various illustrative components and steps have been described above generally in terms of their functionality in order to clearly illustrate this interchangeability of hardware and software. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the implementation. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present application.
The principles and embodiments of the present application are explained herein using specific examples, which are provided only to help understand the method and the core idea of the present application. It will be apparent to those skilled in the art that various changes and modifications can be made in the present invention without departing from the principles of the invention, and these changes and modifications also fall within the scope of the claims of the present application.
It is further noted that, in the present specification, relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
Claims (10)
1. A customer service recommendation method is characterized by comprising the following steps:
acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
inputting the historical consultation data into a preset consultation problem classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories;
receiving a target problem category output by the consultation problem classification model;
and placing the customer service consultation request into a target customer service consultation queue corresponding to the target problem category.
2. The method of claim 1, wherein obtaining historical consulting data from historical consulting sessions with customer services based on the received customer service consulting request comprises:
determining a consultation initiating object according to the received customer service consultation request;
extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
and summarizing all original consultation data in the form of texts to obtain the historical consultation data.
3. The method of claim 2, further comprising:
in response to the absence of a historical consultation session corresponding to the consultation initiating object, determining one of a plurality of alternative question categories as the target question category.
4. The method of claim 2, wherein determining one of the alternative plurality of issue categories as the target issue category comprises:
determining a highest-heat problem category corresponding to the current time period from a plurality of problem categories which are selected;
and determining the highest-heat problem category as the target problem category.
5. The method of any of claims 1-4, further comprising, after placing the customer service consultation request in a target customer service consultation queue corresponding to the target issue category:
controlling the target customer service corresponding to the target customer service consultation queue to process each customer service consultation request in the queue in sequence;
controlling the target customer service to label the target problem category on the consultation content corresponding to the target problem category in the current consultation data after consultation is finished so as to obtain a new training sample;
controlling the target customer service to label the consultation contents which do not correspond to the target problem category in the current consultation data after consultation is finished, so as to obtain a new negative training sample;
and continuously iterating and updating the consultation problem classification model by using the new positive training sample and the new negative training sample.
6. A customer service recommendation device, comprising:
the historical consultation data acquisition unit is configured for acquiring historical consultation data from a historical consultation session with customer service according to the received customer service consultation request;
a consultation problem classification calling model configured to input the historical consultation data into a preset consultation problem classification model; the consultation problem classification model is used for representing the corresponding relation between different historical consultation data and the belonged problem categories;
a target problem category receiving unit configured to receive a target problem category output by the consultation problem classification model;
a targeted queuing unit configured to place the customer service consultation request in a target customer service consultation queue corresponding to the target problem category.
7. The apparatus of claim 6, wherein the historical consulting data obtaining unit is further configured to:
determining a consultation initiating object according to the received customer service consultation request;
extracting at least one kind of original consultation data including text, image and voice from the historical consultation session between the consultation initiating object and the customer service;
converting the original consultation data expressed in the form of images into original consultation data expressed in the form of texts by an optical character recognition technology;
converting the original consultation data expressed in the form of voice into original consultation data expressed in the form of text by a voice recognition technology;
and summarizing all original consultation data in the form of texts to obtain the historical consultation data.
8. The apparatus of claim 7, further comprising:
a first consultation target problem category determining unit configured to determine one of a plurality of candidate problem categories as the target problem category in response to an absence of a historical consultation session corresponding to the consultation initiating object.
9. An electronic device, comprising:
a memory for a computer program;
a processor adapted to perform the steps of the method of customer service recommendation of any of claims 1-5 when executing the computer program stored on the memory.
10. A readable storage medium, characterized in that the readable storage medium has stored thereon a computer program which, when being executed by a processor, is adapted to carry out the steps of the service recommendation method according to any one of claims 1 to 5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306314.0A CN114036379A (en) | 2021-11-05 | 2021-11-05 | Customer service recommendation method and device, electronic equipment and readable storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111306314.0A CN114036379A (en) | 2021-11-05 | 2021-11-05 | Customer service recommendation method and device, electronic equipment and readable storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114036379A true CN114036379A (en) | 2022-02-11 |
Family
ID=80142956
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111306314.0A Pending CN114036379A (en) | 2021-11-05 | 2021-11-05 | Customer service recommendation method and device, electronic equipment and readable storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114036379A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116545970A (en) * | 2023-05-08 | 2023-08-04 | 阿里巴巴（中国）有限公司 | Consultation queue display method and device and electronic equipment |
| CN118396637A (en) * | 2024-06-27 | 2024-07-26 | 江西科技学院 | Online consultation method and system |
Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512153A (en) * | 2014-10-20 | 2016-04-20 | 中国电信股份有限公司 | Method and device for service provision of online customer service system, and system |
| CN110909148A (en) * | 2019-12-17 | 2020-03-24 | 中国银行股份有限公司 | Customer service assisting method and device |
| WO2021174840A1 (en) * | 2020-03-03 | 2021-09-10 | 平安科技（深圳）有限公司 | Voice flow setting method and apparatus, computer device and storage medium |
-
2021
- 2021-11-05 CN CN202111306314.0A patent/CN114036379A/en active Pending
Patent Citations (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN105512153A (en) * | 2014-10-20 | 2016-04-20 | 中国电信股份有限公司 | Method and device for service provision of online customer service system, and system |
| CN110909148A (en) * | 2019-12-17 | 2020-03-24 | 中国银行股份有限公司 | Customer service assisting method and device |
| WO2021174840A1 (en) * | 2020-03-03 | 2021-09-10 | 平安科技（深圳）有限公司 | Voice flow setting method and apparatus, computer device and storage medium |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116545970A (en) * | 2023-05-08 | 2023-08-04 | 阿里巴巴（中国）有限公司 | Consultation queue display method and device and electronic equipment |
| CN118396637A (en) * | 2024-06-27 | 2024-07-26 | 江西科技学院 | Online consultation method and system |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106503236B (en) | Artificial intelligence based problem classification method and device | |
| CN109460514B (en) | Method and device for pushing information | |
| CN112365894B (en) | AI-based composite voice interaction method and device and computer equipment | |
| CN110364146B (en) | Speech recognition method, speech recognition device, speech recognition apparatus, and storage medium | |
| EP3579140A1 (en) | Method and apparatus for processing video | |
| CN108153800B (en) | Information processing method, information processing apparatus, and recording medium | |
| CN110110038B (en) | Telephone traffic prediction method, device, server and storage medium | |
| CN109919244B (en) | Method and apparatus for generating a scene recognition model | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| CN114036379A (en) | Customer service recommendation method and device, electronic equipment and readable storage medium | |
| CN113360622B (en) | User dialogue information processing method and device and computer equipment | |
| CN107784033B (en) | Method and device for recommending based on session | |
| CN112185358A (en) | Intention recognition method, model training method, device, equipment and medium | |
| CN110597965B (en) | Emotion polarity analysis method and device for article, electronic equipment and storage medium | |
| CN116595150A (en) | Dialogue recommendation method, device, equipment and storage medium | |
| CN113342942B (en) | Corpus automatic acquisition method and device, computer equipment and storage medium | |
| CN116050405A (en) | Text processing, question-answer text processing and text processing model training method | |
| CN111611365A (en) | Flow control method, device, equipment and storage medium of dialog system | |
| CN115955452A (en) | 5G message pushing method and device based on multi-turn conversation intention recognition | |
| CN110047473B (en) | Man-machine cooperative interaction method and system | |
| CN111540355A (en) | Personalized setting method and device based on voice assistant | |
| CN113806501B (en) | Training method of intention recognition model, intention recognition method and equipment | |
| CN106371905B (en) | Application program operation method and device and server | |
| CN110263135B (en) | Data exchange matching method, device, medium and electronic equipment | |
| CN111931494A (en) | Method, apparatus, electronic device, and medium for generating prediction information |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination |