CN112632245A - Intelligent customer service distribution method and device, computer equipment and storage medium - Google Patents
Intelligent customer service distribution method and device, computer equipment and storage medium Download PDFInfo
- Publication number
- CN112632245A CN112632245A CN202011511938.1A CN202011511938A CN112632245A CN 112632245 A CN112632245 A CN 112632245A CN 202011511938 A CN202011511938 A CN 202011511938A CN 112632245 A CN112632245 A CN 112632245A
- Authority
- CN
- China
- Prior art keywords
- customer service
- session
- session message
- message
- intelligent customer
- 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
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/33—Querying
- G06F16/332—Query formulation
- G06F16/3329—Natural language query formulation or dialogue systems
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Engineering & Computer Science (AREA)
- Artificial Intelligence (AREA)
- Mathematical Physics (AREA)
- Computational Linguistics (AREA)
- General Physics & Mathematics (AREA)
- Human Computer Interaction (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Health & Medical Sciences (AREA)
- Audiology, Speech & Language Pathology (AREA)
- General Health & Medical Sciences (AREA)
- Information Transfer Between Computers (AREA)
Abstract
The application relates to the field of business process optimization and discloses a method and a device for intelligently distributing customer service, computer equipment and a storage medium, wherein the method comprises the following steps: receiving a session message sent by a user, performing scene analysis on the session message, and determining whether the session message is related to a current service scene; when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message; determining whether customer service data corresponding to the participatory speech meaning exists in a database of the intelligent customer service based on the session semantics; if the intelligent customer service database does not have customer service data corresponding to the participatory meaning, distributing the session message to the manual customer service; and if the intelligent customer service database has customer service data corresponding to the participatory meaning, distributing the session message to the intelligent customer service. And the corresponding customer service type is selected according to the conversation of the user, so that the time of the user is saved, and the user experience is improved.
Description
Technical Field
The present application relates to the field of process optimization technologies, and in particular, to a method and an apparatus for intelligently allocating customer services, a computer device, and a storage medium.
Background
With the rapid development of the internet, a large amount of services begin to be processed on line, and users can not handle related services for business. In order to solve various problems generated in the service handling process of a user and improve the on-line service handling experience of the user, a large number of network customer services are currently set to solve the problems of the user, and the network customer services comprise manual customer services and AI customer services. When receiving the conversation of the user, the AI customer service answers the question of the user, and when the user thinks that the AI customer service cannot answer the question, the user can select to switch to the manual customer service.
The existing customer service process mostly adopts the method to reduce the workload of manual customer service. However, in the actual use process, the time spent by the user is often too long in the process from the AI customer service answer to the manual customer service answer, and the user experience is poor.
Therefore, how to select the corresponding customer service type according to the session of the user becomes a problem to be solved urgently.
Disclosure of Invention
The application provides a method and a device for intelligently distributing customer service, computer equipment and a storage medium, so that the corresponding customer service type is selected according to a session of a user, the time of the user is saved, and the user experience is improved.
In a first aspect, the present application provides an intelligent customer service distribution method, including:
receiving a session message sent by a user, performing scene analysis on the session message, and determining whether the session message is related to a current service scene; when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message; determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics; if the intelligent customer service database does not have customer service data corresponding to the session semantics, distributing the session message to an artificial customer service, and processing the session message by the artificial customer service; and if the intelligent customer service database has customer service data corresponding to the session semantics, distributing the session message to the intelligent customer service, and processing the session message by the intelligent customer service.
In a second aspect, the present application further provides an intelligent customer service distribution device, including:
the scene analysis module is used for receiving the session message sent by the user, carrying out scene analysis on the session message and determining whether the session message is related to the current service scene; the semantic analysis module is used for performing semantic analysis on the session message when the session message is related to the current service scene to obtain session semantics corresponding to the session message; the data determining module is used for determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics; the artificial distribution module is used for distributing the session message to an artificial customer service and processing the session message by the artificial customer service if the customer service data corresponding to the session semantics do not exist in the database of the intelligent customer service; and the intelligent distribution module is used for distributing the session message to the intelligent customer service and processing the session message by the intelligent customer service if the customer service data corresponding to the session semantics exist in the database of the intelligent customer service.
In a third aspect, the present application further provides a computer device, characterized in that the computer device comprises a memory and a processor; the memory is used for storing a computer program; the processor is used for executing the computer program and realizing the intelligent customer service distribution method when the computer program is executed.
In a fourth aspect, the present application further provides a computer-readable storage medium, which stores a computer program, and when the computer program is executed by a processor, the computer program causes the processor to implement the intelligent customer service distribution method as described above.
The application discloses a method, a device, computer equipment and a storage medium for intelligently distributing customer service, which determine whether a session message is related to a current service scene by receiving the session message and carrying out scene analysis on the session message; when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message; determining whether customer service data corresponding to the participatory speech meaning exists in a database of the intelligent customer service based on the session semantics; if the intelligent customer service database does not have customer service data corresponding to the participatory meaning, the session message is forwarded to the artificial customer service; and if the customer service data corresponding to the participatory meaning exists in the database of the intelligent customer service, forwarding the session message to the intelligent customer service, and processing the session message by the intelligent customer service. Before the session message is sent to the response server, the forwarding module distributes the most appropriate customer service type to the user sending the session message through scene analysis and semantic analysis of the session message, so that the time actually required by the user to solve the problem is reduced, the user experience is improved, and the efficiency of the response server in processing the session message is also improved.
Drawings
In order to more clearly illustrate the technical solutions of the embodiments of the present application, the drawings needed to be used in the description of the embodiments are briefly introduced below, and it is obvious that the drawings in the following description are some embodiments of the present application, and it is obvious for those skilled in the art to obtain other drawings based on these drawings without creative efforts.
FIG. 1 is a block diagram schematically illustrating a computer device according to an embodiment of the present disclosure;
FIG. 2 is a schematic flow chart of an intelligent customer service distribution method provided in an embodiment of the present application;
fig. 3 is a schematic block diagram of an intelligent customer service distribution device according to an embodiment of the present disclosure.
Detailed Description
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, but not all, embodiments of the present application. 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.
The flow diagrams depicted in the figures are merely illustrative and do not necessarily include all of the elements and operations/steps, nor do they necessarily have to be performed in the order depicted. For example, some operations/steps may be decomposed, combined or partially combined, so that the actual execution sequence may be changed according to the actual situation.
It is to be understood that the terminology used in the description of the present application herein is for the purpose of describing particular embodiments only and is not intended to be limiting of the application. As used in the specification of the present application and the appended claims, the singular forms "a," "an," and "the" are intended to include the plural forms as well, unless the context clearly indicates otherwise.
It should also be understood that the term "and/or" as used in this specification and the appended claims refers to and includes any and all possible combinations of one or more of the associated listed items.
The embodiment of the application provides a method and a device for intelligently distributing customer service, computer equipment and a storage medium. The intelligent customer service distribution method can be used for distributing proper customer service for the user, so that the time for solving the problem of the user is saved, the user experience is improved, and the processing efficiency of the session message is improved.
Some embodiments of the present application will be described in detail below with reference to the accompanying drawings. The embodiments described below and the features of the embodiments can be combined with each other without conflict.
As shown in fig. 1, fig. 1 is a schematic block diagram of a structure of a computer device provided in an embodiment of the present application.
Referring to fig. 1, the computer device includes a processor, a memory and a network interface connected by a system bus, wherein the memory may include a storage medium and an internal memory.
The storage medium may store an operating system and a computer program. The computer program includes program instructions that, when executed, cause a processor to perform any one of the intelligent customer service distribution methods.
The processor is used for providing calculation and control capability and supporting the operation of the whole computer equipment.
The internal memory provides an environment for the execution of a computer program on a storage medium, which when executed by the processor causes the processor to perform any of the intelligent customer service distribution methods.
The network interface is used for network communication, such as sending assigned tasks and the like. Those skilled in the art will appreciate that the architecture shown in fig. 1 is merely a block diagram of some of the structures associated with the disclosed aspects and is not intended to limit the computing devices to which the disclosed aspects apply, as particular computing devices may include more or less components than those shown, or may combine certain components, or have a different arrangement of components.
It should be understood that the Processor may be a Central Processing Unit (CPU), and the Processor may be other general purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field Programmable Gate Arrays (FPGAs) or other Programmable logic devices, discrete Gate or transistor logic devices, discrete hardware components, etc. Wherein a general purpose processor may be a microprocessor or the processor may be any conventional processor or the like.
Wherein, in one embodiment, the processor is configured to execute a computer program stored in the memory to implement the steps of:
receiving a session message sent by a user, performing scene analysis on the session message, and determining whether the session message is related to a current service scene; when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message; determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics; if the intelligent customer service database does not have customer service data corresponding to the session semantics, distributing the session message to an artificial customer service, and processing the session message by the artificial customer service; and if the intelligent customer service database has customer service data corresponding to the session semantics, distributing the session message to the intelligent customer service, and processing the session message by the intelligent customer service.
In one embodiment, the processor is configured to implement:
acquiring response content sent by the artificial customer service based on the session message; performing context recognition on the response content, and determining whether wrongly written words exist in the response content; and if the response content is determined to have wrongly written characters, suspending sending the response content and reminding the manual customer service.
In one embodiment, the processor is configured to implement:
when the conversation message comprises a conversation image, identifying the content of the conversation image, and determining whether the conversation image is an illegal image; and if the conversation image is not an illegal image, executing the step of performing scene analysis on the conversation message.
In one embodiment, the processor is configured to implement:
and if the session message is not related to the current service scene, sending a prompt to the user and terminating the distribution of the session message.
In one embodiment, the processor, prior to effecting said distributing said session message to said smart customer service, is adapted to effect:
performing emotion analysis on the conversation message to obtain an emotion type corresponding to the conversation message; if the emotion type is a preset type, distributing the conversation message to a manual customer service; and if the emotion type is not the preset type, distributing the conversation message to the intelligent customer service.
In one embodiment, the processor, when implementing the scene analysis on the conversation message, is configured to implement:
preprocessing the conversation message to obtain a preprocessed conversation message, wherein the preprocessing comprises at least one of word segmentation processing, synonym adding, stop word removing and punctuation mark removing; and carrying out scene analysis on the preprocessed session message by utilizing a pre-trained scene recognition model.
In one embodiment, the processor, when performing the semantic analysis on the conversation message, is configured to perform:
and converting the message content of the session message into text data, and performing semantic analysis on the text data.
In one embodiment, the computer device may include a forwarding module and a response module, the response module including an intelligent response submodule and a manual response submodule. The forwarding module is used for distributing and forwarding the session message initiated by the user, and the response module is used for making a response according to the received session message.
In the specific implementation process, when a user initiates a consultation session through a user side, a session message is firstly sent to a forwarding module, the forwarding module performs scene analysis on the session message, and if the scene of the session message is different from the current scene of the forwarding module, the processing of the session message is terminated, the session message is not forwarded at all, and the user is prompted.
And if the conversation message is the same as the current scene of the forwarding module, performing semantic analysis on the conversation message to obtain conversation semantics, and determining whether the intelligent customer service database has customer service data corresponding to the conversation semantics according to the conversation semantics.
If the intelligent customer service database does not have the customer service data corresponding to the session semantics, the intelligent customer service determines that the intelligent customer service cannot process the session message, and forwards the session message to a manual response submodule in a response module for response.
If the intelligent customer service database has customer service data corresponding to the session semantics, the intelligent customer service determines that the intelligent customer service can process the session message, and forwards the session message to an intelligent response submodule in a response module for response.
The intelligent response sub-module is connected with the intelligent customer service, and when the forwarding module forwards the session message to the intelligent customer service module, the intelligent customer service responds to the session message.
The manual response sub-module is connected with the manual customer service, and when the forwarding module forwards the session message to the manual customer service module, the manual customer service responds to the session message.
When the intelligent response sub-module receives the session message, the intelligent response sub-module calls the customer service data related to the session message from the database of the intelligent customer service, makes a response based on the customer service data, and directly sends the response to the user side.
When the manual response sub-module receives the session message, the manual response is carried out based on the content of the session message, the response is sent to the forwarding module, the forwarding module checks the response content, and the response content is sent to the user side after being checked to be correct.
The step of checking the response content by the forwarding module may specifically be determining whether a wrongly written word exists in the response content. If the wrongly written characters exist, the sending of the response content is stopped, manual customer service is prompted to modify the wrongly written characters in the response content, and the wrongly written characters are sent again after modification. If there is no wrongly written character, the response content is directly sent to the user terminal.
Referring to fig. 2, fig. 2 is a schematic flowchart of an intelligent customer service distribution method according to an embodiment of the present application. The intelligent customer service distribution method performs scene analysis and semantic analysis on the session message, selects an appropriate customer service type according to the content of the session message, and improves the user experience and the efficiency of processing the session message.
As shown in fig. 2, the intelligent customer service distribution method specifically includes: step S201 to step S205.
S201, receiving a session message sent by a user, performing scene analysis on the session message, and determining whether the session message is related to a current service scene.
When a user needs to perform problem consultation on customer service, a consultation session is initiated through a user side, the user side sends session messages to a forwarding module, and the forwarding module performs scene analysis on the session messages after receiving the session messages, so that whether the session messages are related to the current service scene or not is identified.
By carrying out scene analysis on the session message, occupation of meaningless and unsolved session messages on customer service resources is reduced.
The content of the session message may be various, for example, it may be a picture, a text, or various formats such as voice or video. And determining whether the session message is related to the current service scene, mainly determining whether the content of the session message sent by the user is the content of the service scene of the current customer service system.
For example, if the current service scenario is a credit card claim scenario, the content of the session message sent by the user is a credit card transaction condition, and the session message is analyzed to determine that the session message is related to the current service scenario. If the current service scene is a credit card application scene, the content of the session message sent by the user is the problem of the deposit card periodically stored, and the session message is analyzed to determine that the session message is irrelevant to the current service scene if the scene of the session message is considered to be a deposit card deposit scene.
In one embodiment, the intelligent customer service distribution method comprises the following steps: when the conversation message comprises a conversation image, identifying the content of the conversation image, and determining whether the conversation image is an illegal image; and if the conversation image is not an illegal image, executing the step of performing scene analysis on the conversation message.
When the conversation message comprises the conversation image, namely when the content in the conversation message is the image, the content of the conversation image is identified, and whether the conversation image is an illegal image is identified. The illegal image is a pornographic image, a horror image, an image prohibited from being released by a country, or the like.
In the specific implementation process, a sensitive word library, a font library and a color library can be established. And respectively combining the materials in the sensitive word library, the font library and the color library to obtain a sensitive image corresponding to each combination.
And for each image in the session message, matching the sensitive image with each image in the session message, determining the matching degree of the image and the sensitive image, considering the image as an illegal image when the matching degree is greater than a preset threshold value, determining that the session message is abnormal, terminating sending the session message, not forwarding the session message, and prompting a user.
And if the image is confirmed not to be the illegal image, performing scene analysis on the session message so as to determine whether the session message is related to the current service scene.
In an embodiment, the step of performing the scene analysis on the session message specifically includes: preprocessing the session message to obtain a preprocessed session message; and carrying out scene analysis on the preprocessed session message by utilizing a pre-trained scene recognition model.
Wherein the preprocessing comprises at least one of word segmentation processing, synonym adding, stop word removing and punctuation mark removing. Preprocessing the session message to obtain a preprocessed session message, inputting the preprocessed session message into a pre-trained scene recognition model to perform scene analysis, and determining a scene corresponding to the session message.
When the scene recognition model is trained, firstly, the training data is labeled, and the scene type corresponding to each training data is labeled to obtain sample data. And then preprocessing the sample data, wherein the preprocessing comprises word segmentation, synonym addition, stop word removal, punctuation removal and the like, for example, a jieba word segmenter can be used for segmenting the sample data, then stop words such as's','s' and the like in the sample data are removed, synonyms are added, for example, the synonym of the air ticket is an airplane ticket, and the synonym of the purchase is a purchase and the like.
For the preprocessed sample data, converting the preprocessed sample data into word vectors by using a word2vec model, preprocessing the word vectors, wherein the preprocessing comprises word mapping and length filling, the word mapping refers to compiling words into integer series, the filling of sentence length refers to the fact that the length filling of each text is consistent, the number of text words with the maximum length is used as the maximum length, and the short text is filled, so that the length filling of each text is consistent.
And inputting the preprocessed word vector into a textCNN neural network model, and training the neural network model to obtain a scene recognition model.
S202, when the session message is relevant to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message.
And when the session message is related to the current service scene, performing semantic analysis on the session content in the session message to obtain the session semantics corresponding to the session message, so as to distribute corresponding types of customer service to the user according to the session semantics. The session semantics refer to the problem that the user actually wants to solve or the content of communication when sending the session message.
In an embodiment, the semantic analyzing the conversation message includes: and converting the message content of the session message into text data, and performing semantic analysis on the text data.
Because the message content of the session message can have various formats, the message content of the session message can be converted into text data during semantic analysis, and the text data is used for semantic analysis. In the semantic analysis, for example, a natural language processing technique may be used to perform the semantic analysis.
When the message content of the conversation message is an image, firstly, extracting characters from the image to obtain the characters included in the image, namely text data corresponding to the image, and then, performing semantic analysis according to the text data.
When the message content of the conversation message is voice, voice recognition is firstly carried out on the voice, the voice is converted into a text by utilizing a voice recognition technology, and then semantic analysis is carried out on the text.
In one embodiment, the intelligent customer service distribution method comprises the following steps: and if the session message is not related to the current service scene, sending a prompt to the user and terminating the distribution of the session message.
When the session message is determined not to be related to the current service scene, a prompt can be sent to the user in a prompt box mode to remind the user that the customer service in the current service scene cannot answer the question, the sending of the session message is terminated, and the session message is not sent to any type of customer service.
In the specific implementation process, in the prompt sent to the user, in addition to reminding the user that the customer service in the current service scene cannot answer the question, the user can be informed of the service scene customer service that may solve the question, so that the user can select the customer service in other service scenes for service.
In an embodiment, if it cannot be determined whether the session message is related to the current service scenario, the session message is forwarded to the artificial customer service, and the artificial customer service replies based on the session message.
Because the content of the session message may be of various types, if the session message cannot be correctly identified, the scene analysis of the session message may not be performed normally, and it may not be determined whether the session message is related to the current service scene.
When the conversation message is not related to the current service scene, the conversation message is directly forwarded to the artificial customer service part, and the artificial customer service part replies based on the conversation message, so that the user experience is improved.
S203, determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics.
After the session message is subjected to semantic analysis to obtain session semantics, the customer service data in the database of the intelligent customer service can be retrieved based on the session semantics, so that whether the customer service data corresponding to the participatory speech semantics exist in the database of the intelligent customer service is determined.
The customer service data included in the database of the intelligent customer service may be pre-stored, and include some common problems and responses of the common problems, and contents such as commonly used phrases. The common terms may be, for example, please wait, don't mean, sorry, etc.
If the intelligent customer service database has customer service data corresponding to the participatory speech meaning, the intelligent customer service is considered to be capable of processing the conversation message; and if the intelligent customer service database does not have the customer service data corresponding to the participatory meaning, the intelligent customer service is considered to be incapable of processing the conversation message.
In one embodiment, before said distributing said session message to said smart customer service, said method comprises: performing emotion analysis on the conversation message to obtain an emotion type corresponding to the conversation message; if the emotion type is a preset type, distributing the conversation message to a manual customer service; and if the emotion type is not the preset type, distributing the conversation message to the intelligent customer service.
And performing emotion analysis on the conversation message to obtain the emotion type when the user initiates the conversation message. The emotion types may be of various types such as positive, negative, and neutral. Different emotion analysis methods can be used for emotion analysis for different types of conversation messages.
For example, for a text conversation message, the emotion type may be determined according to the text content, and when the text such as urgent or urgent appears in the text content many times, the emotion type is considered to be urgent.
For conversational messages of the speech type, the emotion recognition model can be trained to recognize the emotion type. And acquiring voice data of the user under different emotions, labeling the voice data, and constructing a training set. And training the training set by using the LSTM to obtain an emotion recognition model, and inputting the speech conversation message into the trained emotion recognition model to obtain the emotion type.
The emotion recognition model comprises a convolutional layer, a full connection layer and a Softmax layer. And (5) performing feature extraction on the convolutional layer, and outputting the classification probability through the full connection layer and the Softmax layer to obtain the emotion type.
The preset type may be a preset type, such as a negative type, that needs to be switched to manual customer service. And when emotion analysis is carried out on the conversation message and the obtained emotion type belongs to a preset type, the conversation message is considered to be required to be processed by manual customer service.
Therefore, when the emotion type obtained by analysis is the preset type, the conversation message is forwarded to the artificial customer service, and the artificial customer service processes the conversation message. And when the emotion type obtained by analysis is not the preset type, forwarding the conversation message to the intelligent customer service, and processing the conversation message by the intelligent customer service.
S204, if the intelligent customer service database does not have customer service data corresponding to the session semantics, distributing the session message to an artificial customer service, and processing the session message by the artificial customer service.
If the intelligent customer service database does not have customer service data corresponding to the participatory meaning, the intelligent customer service determines that the intelligent customer service cannot process the session message at the moment, the session message is forwarded to the artificial customer service, and the artificial customer service processes the session message.
In one embodiment, the intelligent customer service distribution method further includes: acquiring response content sent by the artificial customer service based on the session message; performing context recognition on the response content, and determining whether wrongly written words exist in the response content; and if the response content is determined to have wrongly written characters, suspending sending the response content and reminding the manual customer service.
After the forwarding module forwards the session message to the artificial customer service, the artificial customer service makes a response based on the session message, sends the response content to the forwarding module, and forwards the response content to the user by the forwarding module.
After the forwarding module obtains the response content sent by the artificial customer service, the context of the response content is identified to determine whether wrongly written characters exist in the response content sent by the artificial customer service, if the wrongly written characters exist in the response content, the sending of the response content is stopped, and the artificial customer service is reminded to modify the wrongly written characters in the response content.
In the specific implementation process, a Chinese word segmentation device can be used for segmenting the response content, errors are detected from two aspects of word granularity and word granularity, and when the errors are detected, the wrongly-written characters are considered to exist in the response content.
In an implementation process, if the forwarding module determines that the wrongly written characters exist in the response content, the wrongly written characters are highlighted so that a human customer service can confirm whether the highlighted part is the wrongly written characters. And if the manual customer service confirms that the highlighted part is not a wrongly written character, directly sending the response content to the user. And if the manual customer service confirms that the highlighted part is wrongly written, stopping sending the response content, and reminding the manual customer service of sending the response content after modification.
S205, if the intelligent customer service database has customer service data corresponding to the session semantics, distributing the session message to the intelligent customer service, and processing the session message by the intelligent customer service.
If the intelligent customer service database has customer service data corresponding to the participatory meaning, the intelligent customer service considers that the intelligent customer service can process the session message at the moment, the session message is forwarded to the intelligent customer service, and the intelligent customer service calls the customer service data in the intelligent customer service database to process the session message.
The intelligent customer service distribution method provided by the embodiment determines whether the session message is related to the current service scene or not by receiving the session message and performing scene analysis on the session message; when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message; determining whether customer service data corresponding to the participatory speech meaning exists in a database of the intelligent customer service based on the session semantics; if the intelligent customer service database does not have customer service data corresponding to the participatory meaning, the session message is forwarded to the artificial customer service; and if the customer service data corresponding to the participatory meaning exists in the database of the intelligent customer service, forwarding the session message to the intelligent customer service, and processing the session message by the intelligent customer service. Before the session message is sent to the response server, the forwarding module distributes the most appropriate customer service type to the user sending the session message through scene analysis and semantic analysis of the session message, so that the time actually required by the user to solve the problem is reduced, the user experience is improved, and the efficiency of the response server in processing the session message is also improved.
Referring to fig. 3, fig. 3 is a schematic block diagram of an intelligent customer service distribution device according to an embodiment of the present application, where the intelligent customer service distribution device is configured to execute the foregoing intelligent customer service distribution method. The intelligent customer service distribution device can be configured in a server or a terminal.
The server may be an independent server or a server cluster. The terminal can be an electronic device such as a mobile phone, a tablet computer, a notebook computer, a desktop computer, a personal digital assistant and a wearable device.
As shown in fig. 3, the intelligent customer service distribution device 300 includes: a scene analysis module 301, a semantic analysis module 302, a data determination module 303, a manual assignment module 304, and an intelligent assignment module 305.
The scene analysis module 301 is configured to receive a session message sent by a user, perform scene analysis on the session message, and determine whether the session message is related to a current service scene.
A semantic analysis module 302, configured to perform semantic analysis on the session message when the session message is related to a current service scenario, so as to obtain a session semantic corresponding to the session message.
And a data determining module 303, configured to determine, based on the session semantics, whether customer service data corresponding to the session semantics exists in a database of the intelligent customer service.
And an artificial distribution module 304, configured to distribute the session message to an artificial customer service if there is no customer service data corresponding to the session semantics in the database of the intelligent customer service, and process the session message by the artificial customer service.
An intelligent distribution module 305, configured to distribute the session message to the intelligent customer service and process the session message by the intelligent customer service if the customer service data corresponding to the session semantics exists in the database of the intelligent customer service.
It should be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working processes of the customer service intelligent allocation apparatus and each module described above may refer to the corresponding processes in the foregoing customer service intelligent allocation method embodiment, and are not described herein again.
The intelligent customer service distribution device described above may be implemented in the form of a computer program that can be run on a computer device as shown in fig. 1.
The embodiment of the application further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, the computer program comprises program instructions, and the processor executes the program instructions to realize any intelligent customer service distribution method provided by the embodiment of the application.
The computer-readable storage medium may be an internal storage unit of the computer device described in the foregoing embodiment, for example, a hard disk or a memory of the computer device. The computer readable storage medium may also be an external storage device of the computer device, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), etc. provided on the computer device, and the storage medium may be nonvolatile or volatile.
While the invention has been described with reference to specific embodiments, the scope of the invention is not limited thereto, and those skilled in the art can easily conceive various equivalent modifications or substitutions within the technical scope of the invention. Therefore, the protection scope of the present application shall be subject to the protection scope of the claims.
Claims (10)
1. An intelligent customer service distribution method, characterized in that the method comprises:
receiving a session message sent by a user, performing scene analysis on the session message, and determining whether the session message is related to a current service scene;
when the session message is related to the current service scene, performing semantic analysis on the session message to obtain session semantics corresponding to the session message;
determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics;
if the intelligent customer service database does not have customer service data corresponding to the session semantics, distributing the session message to an artificial customer service, and processing the session message by the artificial customer service;
and if the intelligent customer service database has customer service data corresponding to the session semantics, distributing the session message to the intelligent customer service, and processing the session message by the intelligent customer service.
2. The intelligent customer service distribution method according to claim 1, wherein the method comprises:
acquiring response content sent by the artificial customer service based on the session message;
performing context recognition on the response content, and determining whether wrongly written words exist in the response content;
and if the response content is determined to have wrongly written characters, suspending sending the response content and reminding the manual customer service.
3. The intelligent customer service distribution method according to claim 1, wherein the method comprises:
when the conversation message comprises a conversation image, identifying the content of the conversation image, and determining whether the conversation image is an illegal image;
and if the conversation image is not an illegal image, executing the step of performing scene analysis on the conversation message.
4. The intelligent customer service distribution method according to claim 1, wherein the method comprises:
and if the session message is not related to the current service scene, sending a prompt to the user and terminating the distribution of the session message.
5. The intelligent customer service distribution method according to claim 1, wherein prior to said distributing said session message to said intelligent customer service, said method comprises:
performing emotion analysis on the conversation message to obtain an emotion type corresponding to the conversation message;
if the emotion type is a preset type, distributing the conversation message to a manual customer service;
and if the emotion type is not the preset type, distributing the conversation message to the intelligent customer service.
6. The intelligent customer service distribution method according to claim 1, wherein the performing a scene analysis on the conversation message comprises:
preprocessing the conversation message to obtain a preprocessed conversation message, wherein the preprocessing comprises at least one of word segmentation processing, synonym adding, stop word removing and punctuation mark removing;
and carrying out scene analysis on the preprocessed session message by utilizing a pre-trained scene recognition model.
7. The intelligent customer service distribution method according to claim 1, wherein the semantic analysis of the session message comprises:
and converting the message content of the session message into text data, and performing semantic analysis on the text data.
8. An intelligent customer service distribution device, the device comprising:
the scene analysis module is used for receiving the session message sent by the user, carrying out scene analysis on the session message and determining whether the session message is related to the current service scene;
the semantic analysis module is used for performing semantic analysis on the session message when the session message is related to the current service scene to obtain session semantics corresponding to the session message;
the data determining module is used for determining whether customer service data corresponding to the session semantics exist in a database of the intelligent customer service based on the session semantics;
the artificial distribution module is used for distributing the session message to an artificial customer service and processing the session message by the artificial customer service if the customer service data corresponding to the session semantics do not exist in the database of the intelligent customer service;
and the intelligent distribution module is used for distributing the session message to the intelligent customer service and processing the session message by the intelligent customer service if the customer service data corresponding to the session semantics exist in the database of the intelligent customer service.
9. A computer device, wherein the computer device comprises a memory and a processor;
the memory is used for storing a computer program;
the processor, configured to execute the computer program and to implement the intelligent customer service allocation method according to any one of claims 1 to 7 when executing the computer program.
10. A computer-readable storage medium, characterized in that it stores a computer program which, when executed by a processor, causes the processor to carry out the intelligent customer service distribution method according to any one of claims 1 to 7.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011511938.1A CN112632245A (en) | 2020-12-18 | 2020-12-18 | Intelligent customer service distribution method and device, computer equipment and storage medium |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202011511938.1A CN112632245A (en) | 2020-12-18 | 2020-12-18 | Intelligent customer service distribution method and device, computer equipment and storage medium |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN112632245A true CN112632245A (en) | 2021-04-09 |
Family
ID=75317764
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202011511938.1A Pending CN112632245A (en) | 2020-12-18 | 2020-12-18 | Intelligent customer service distribution method and device, computer equipment and storage medium |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN112632245A (en) |
Cited By (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113112281A (en) * | 2021-04-16 | 2021-07-13 | 上海岐力电子商务有限公司 | Customer service training method and customer service system based on artificial intelligence |
| CN113159901A (en) * | 2021-04-29 | 2021-07-23 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease service session |
| CN113345468A (en) * | 2021-05-25 | 2021-09-03 | 平安银行股份有限公司 | Voice quality inspection method, device, equipment and storage medium |
| CN113505268A (en) * | 2021-07-07 | 2021-10-15 | 中国工商银行股份有限公司 | Interactive processing method and device |
-
2020
- 2020-12-18 CN CN202011511938.1A patent/CN112632245A/en active Pending
Cited By (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN113112281A (en) * | 2021-04-16 | 2021-07-13 | 上海岐力电子商务有限公司 | Customer service training method and customer service system based on artificial intelligence |
| CN113159901A (en) * | 2021-04-29 | 2021-07-23 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease service session |
| CN113159901B (en) * | 2021-04-29 | 2024-06-04 | 天津狮拓信息技术有限公司 | Method and device for realizing financing lease business session |
| CN113345468A (en) * | 2021-05-25 | 2021-09-03 | 平安银行股份有限公司 | Voice quality inspection method, device, equipment and storage medium |
| CN113505268A (en) * | 2021-07-07 | 2021-10-15 | 中国工商银行股份有限公司 | Interactive processing method and device |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN109587360B (en) | Electronic device, method for coping with tactical recommendation, and computer-readable storage medium | |
| CN112632245A (en) | Intelligent customer service distribution method and device, computer equipment and storage medium | |
| CN109697291B (en) | Text semantic paragraph recognition method and device | |
| US10049153B2 (en) | Method for dynamically assigning question priority based on question extraction and domain dictionary | |
| CN111428010B (en) | Man-machine intelligent question-answering method and device | |
| CN110597952A (en) | Information processing method, server, and computer storage medium | |
| CN115035538A (en) | Training method of text recognition model, and text recognition method and device | |
| CN113240510B (en) | Abnormal user prediction method, device, equipment and storage medium | |
| CN114757176A (en) | Method for obtaining target intention recognition model and intention recognition method | |
| CN110046648B (en) | Method and device for classifying business based on at least one business classification model | |
| CN113239204A (en) | Text classification method and device, electronic equipment and computer-readable storage medium | |
| CN108366052A (en) | Verify the processing method and system of short message | |
| CN113849474A (en) | Data processing method and device, electronic equipment and readable storage medium | |
| CN113051380A (en) | Information generation method and device, electronic equipment and storage medium | |
| CN109902146A (en) | Credit information acquisition methods, device, terminal and storage medium | |
| CN112951233A (en) | Voice question and answer method and device, electronic equipment and readable storage medium | |
| KR20190074508A (en) | Method for crowdsourcing data of chat model for chatbot | |
| CN107623620B (en) | Processing method of random interaction data, network server and intelligent dialogue system | |
| CN113051381B (en) | Information quality inspection method, information quality inspection device, computer system and computer readable storage medium | |
| CN115620726A (en) | Voice text generation method, and training method and device of voice text generation model | |
| US11947872B1 (en) | Natural language processing platform for automated event analysis, translation, and transcription verification | |
| CN114416986A (en) | Text data cleaning method and device and storage medium | |
| CN109325234B (en) | Sentence processing method, sentence processing device and computer readable storage medium | |
| CN113283218A (en) | Semantic text compression method and computer equipment | |
| CN112632241A (en) | Method, device, equipment and computer readable medium for intelligent conversation |
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 |