CN116466930A

CN116466930A - Construction method, message processing method and device of intelligent customer service model and electronic equipment

Info

Publication number: CN116466930A
Application number: CN202310253946.8A
Authority: CN
Inventors: 赵泽昊; 苏保林; 高执
Original assignee: Netease Hangzhou Network Co Ltd
Current assignee: Netease Hangzhou Network Co Ltd
Priority date: 2023-03-09
Filing date: 2023-03-09
Publication date: 2023-07-21

Abstract

The invention provides a construction method, a message processing method, a device and electronic equipment of an intelligent customer service model, which comprise the following steps: displaying a plurality of candidate functional components through a graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model; determining at least one target functional component in response to a drag operation for the candidate functional component, and generating a flow node based on the target functional component; responding to the association operation aiming at the flow nodes, and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations for the flow nodes; and constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values. The invention can effectively reduce the development difficulty of the intelligent customer service model, reduce the development and update period of the intelligent customer service model, and effectively reduce the labor cost and time cost required by constructing the intelligent customer service model.

Description

Construction method, message processing method and device of intelligent customer service model and electronic equipment

Technical Field

The present invention relates to the field of data processing technologies, and in particular, to a method and an apparatus for constructing an intelligent customer service model, and an electronic device.

Background

At present, the traditional customer service robot construction flow is composed of links such as demand capturing, demand analysis, design, implementation and test by professional developers through code writing according to user demands, so that the release and update period is long, higher service quality cannot be ensured, business demands cannot be flexibly processed, and a large amount of labor cost and time cost are consumed.

Disclosure of Invention

Accordingly, the present invention is directed to a method, a message processing method, a device and an electronic apparatus for constructing an intelligent customer service model, which can effectively reduce the development difficulty of the intelligent customer service model, reduce the development and update period of the intelligent customer service model, and effectively reduce the labor cost and time cost required for constructing the intelligent customer service model.

In a first aspect, an embodiment of the present invention provides a method for constructing an intelligent customer service model, where a graphical user interface is provided through a terminal, including: displaying a plurality of candidate functional components through the graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model; determining at least one target functional component in response to a drag operation for the candidate functional component, and generating a flow node based on the target functional component; responding to the association operation aiming at the flow nodes, and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations directed to the flow nodes; and constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values.

In a second aspect, an embodiment of the present invention further provides a message processing method, where the method is applied to a server, and the method includes: monitoring at least one message to be processed; message processing is carried out on each message to be processed through a pre-built intelligent customer service model, and a response result corresponding to each message to be processed is obtained; the intelligent customer service model is obtained by loading the server from a designated storage area, and is constructed by adopting any one of the construction methods of the intelligent customer service model provided in the first aspect.

In a third aspect, an embodiment of the present invention further provides a device for constructing an intelligent customer service model, where a graphical user interface is provided through a terminal, including: a component display module for displaying a plurality of candidate functional components through the graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model; the node generation module is used for responding to the drag operation aiming at the candidate functional components, determining at least one target functional component and generating a flow node based on the target functional component; the association editing module is used for responding to association operation aiming at the flow nodes and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations directed to the flow nodes; and the model construction module is used for constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values.

In a fourth aspect, an embodiment of the present invention further provides a message processing apparatus, where the apparatus is applied to a server, and the apparatus includes: the message monitoring module is used for monitoring at least one message to be processed; the message processing module is used for processing the message of each message to be processed through a pre-constructed intelligent customer service model to obtain a response result corresponding to each message to be processed; the intelligent customer service model is obtained by loading the server from a designated storage area, and is constructed by adopting any one of the construction methods of the intelligent customer service model provided in the first aspect.

In a fifth aspect, an embodiment of the present invention further provides an electronic device, including a processor and a memory storing computer-executable instructions executable by the processor to implement the method of any one of the first and second aspects.

In a sixth aspect, embodiments of the present invention also provide a computer-readable storage medium storing computer-executable instructions that, when invoked and executed by a processor, cause the processor to implement the method of any one of the first and second aspects.

According to the method, the device and the electronic equipment for constructing the intelligent customer service model, a plurality of candidate functional components are displayed through a graphical user interface, the candidate functional components are used for providing functions required for constructing the intelligent customer service model, then at least one target functional component is determined in response to dragging operation aiming at the candidate functional components, a flow node is generated based on the target functional components, an association relation between the flow nodes is established in response to association operation aiming at the flow node, attribute parameter values of the flow nodes are determined in response to editing operation aiming at the flow node, and finally the intelligent customer service model is constructed based on the association relation between the flow nodes and the attribute parameter values. According to the method, the intelligent customer service model is built in a dragging and arranging mode of the functional components with rich types, so that service scenes can be met rapidly, development difficulty of the intelligent customer service model can be reduced effectively, development and update periods of the intelligent customer service model are shortened, and labor cost and time cost required for building the intelligent customer service model are reduced effectively.

The message processing method, the device and the electronic equipment provided by the embodiment of the invention are applied to a server, when at least one message to be processed is monitored, each message to be processed can be processed through a pre-built intelligent customer service model to obtain a response result corresponding to each message to be processed, the intelligent customer service model is obtained by loading the server from a designated storage area, and the intelligent customer service model is obtained by building the intelligent customer service model. Compared with the prior art that the service quality cannot be guaranteed in manual processing, the method for processing the information by using the intelligent customer service model to answer the information to be processed can remarkably improve the service effect and remarkably reduce the labor cost consumed by processing the information.

Additional features and advantages of the invention will be set forth in the description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The objectives and other advantages of the invention will be realized and attained by the structure particularly pointed out in the written description and claims hereof as well as the appended drawings.

In order to make the above objects, features and advantages of the present invention more comprehensible, preferred embodiments accompanied with figures are described in detail below.

Drawings

In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings that are needed in the description of the embodiments or the prior art will be briefly described, and it is obvious that the drawings in the description below are some embodiments of the present invention, and other drawings can be obtained according to the drawings without inventive effort for a person skilled in the art.

FIG. 1 is a schematic flow chart of a method for constructing an intelligent customer service model according to an embodiment of the present invention;

FIG. 2 is a schematic flow chart of another method for constructing an intelligent customer service model according to an embodiment of the present invention;

FIG. 3 is a dialogue flow chart of an intelligent customer service model according to an embodiment of the present invention;

FIG. 4 is a schematic diagram of a test interface of an intelligent customer service model according to an embodiment of the present invention;

fig. 5 is a schematic diagram of an online dialogue effect of an intelligent customer service model according to an embodiment of the present invention;

FIG. 6 is a schematic diagram of a record query interface of an intelligent customer service model according to an embodiment of the present invention;

FIG. 7 is a schematic diagram of a method for constructing an intelligent customer service model according to an embodiment of the present invention;

fig. 8 is a flow chart of a message processing method according to an embodiment of the present invention;

fig. 9 is a schematic diagram of a message monitoring module according to an embodiment of the present invention;

FIG. 10 is a diagram of a message processing module architecture according to an embodiment of the present invention;

FIG. 11 is a schematic diagram of an intelligent customer service model according to an embodiment of the present invention;

FIG. 12 is a schematic diagram of another intelligent customer service model according to an embodiment of the present invention;

FIG. 13 is a schematic diagram of another intelligent customer service model according to an embodiment of the present invention;

FIG. 14 is a schematic diagram of another intelligent customer service model according to an embodiment of the present invention;

FIG. 15 is a schematic diagram of another intelligent customer service model according to an embodiment of the present invention;

FIG. 16 is a schematic diagram of another intelligent customer service model according to an embodiment of the present invention;

FIG. 17 is a schematic structural diagram of a device for constructing an intelligent customer service model according to an embodiment of the present invention;

fig. 18 is a schematic structural diagram of a message processing apparatus according to an embodiment of the present invention;

fig. 19 is a schematic structural diagram of an electronic device according to an embodiment of the present invention.

Detailed Description

For the purpose of making the objects, technical solutions and advantages of the embodiments of the present invention more apparent, the technical solutions of the present invention will be clearly and completely described in conjunction with the embodiments, and it is apparent that the described embodiments are some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.

At present, a traditional customer service robot construction process requires a large amount of dialogue rule processing logic for traditional developers to realize a dialogue process circulation function of the robot and a user, and requires a corpus with a certain data volume and a high-performance server to train an NLP (Natural Language Processing ) dialogue model of the customer service robot.

Compared with the low code characteristic of the embodiment of the invention, the traditional customer service robot construction scheme has the following technical defects: (1) Professional developers need to develop according to requirements, follow-up version iterative updating, testing and online maintenance are required to be responsible, more labor cost is required to be consumed, the release and update period is long, and business requirements cannot be flexibly processed; (2) In the traditional development process, the conversation process rule of the customer service robot cannot be arranged through dragging, so that a large amount of redundant codes are caused, and a developer cannot focus on the core business logic; the larger the engineering project maintained by the developer is, the more the hidden risk is, and the higher service quality of the customer service robot cannot be ensured; (3) The customer service robot needs to have dialogue capability based on rules and retrieval and also needs to have natural language dialogue capability, so that developers need a large number of corpora and high-performance servers to train NLP dialogue models, and collection of corresponding resources needs high cost.

Based on the above, the embodiment of the invention provides a method, a message processing method, a device and electronic equipment for constructing an intelligent customer service model, which can effectively reduce the development difficulty of the intelligent customer service model, reduce the development and update period of the intelligent customer service model, and effectively reduce the labor cost and time cost required for constructing the intelligent customer service model.

For the convenience of understanding the present embodiment, first, a method for constructing an intelligent customer service model disclosed in the present embodiment of the present invention is described in detail, and a graphical user interface is provided through a terminal, and referring to a schematic flow chart of a method for constructing an intelligent customer service model shown in fig. 1, the method mainly includes steps S102 to S108:

step S102, displaying a plurality of candidate functional components through a graphical user interface; the candidate functional component may be a low-code component, the candidate functional component is used for providing functions required for building an intelligent customer service model, the candidate functional component includes one or more of a message sending component, a problem collecting component, an interface calling component, a branch jumping component, a display component and a code custom component, the message sending component is used for sending rich messages to a client of a user, the problem collecting component is used for collecting user information through formulating problems, the interface calling component is used for sending an HTTP (Hyper Text Transfer Protocol, a hypertext transfer protocol) request to call an API (Application Program Interface, an application programming interface) interface of other systems, the branch jumping component is used for setting one or more conditions and jumping to a flow node of a corresponding branch after the conditions are met, the display component supports template rendering and cyclic rendering, and the code custom component supports engineers to inject custom codes.

Step S104, at least one target functional component is determined in response to the drag operation for the candidate functional components, and a flow node is generated based on the target functional component. The terminal is configured with a rule arrangement module, and the rule arrangement module is used for constructing an intelligent customer service model, and the corresponding flow node can be generated by dragging the low code component into the rule arrangement module. Optionally, the rendering engine used in the embodiment of the present invention is a reactiflow, and the drag function may be implemented by draft.

Step S106, establishing an association relationship between the flow nodes in response to the association operation for the flow nodes; and determining attribute parameter values for the flow node in response to the editing operation for the flow node. The association relationship may also be referred to as a circulation relationship.

In one embodiment, when a triggering operation aiming at an association control in a graphical user interface is monitored, an association function is determined to be started, a selecting operation aiming at flow nodes is monitored, and an association relation between the two selected flow nodes is established. Optionally, the connection (edge) between two flow nodes is established to represent the association relationship between the two flow nodes.

In one embodiment, when an editing operation (such as double clicking on a flow node) for the flow node is monitored, an attribute item corresponding to the flow node can be displayed through a graphical user interface, and an attribute parameter value corresponding to each target attribute item is determined in response to a modification operation for at least one target attribute item in the attribute items. By way of example, the modifiable attribute items in the question collection component may include one or more of "questions posed to the user", "collect user entered validation rules", "allow user retry times", "save user entered variable names", and the modifiable attribute items in the interface invocation component may include one or more of "flow node names", "target URL (uniform resource locator, uniform resource location system) addresses", "HTTP request Header parameters", "HTTP request Body parameters", "request results to save json data".

And S108, constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values. In one embodiment, when the determining operation for the association relationship and the attribute parameter value is monitored, the intelligent customer service model is obtained.

According to the method for constructing the intelligent customer service model, the intelligent customer service model is constructed in a dragging and arranging mode of the functional components with rich types, so that service scenes can be met rapidly, development difficulty of the intelligent customer service model can be reduced effectively, development and update cycles of the intelligent customer service model are shortened, and labor cost and time cost required for constructing the intelligent customer service model are reduced effectively.

In one embodiment, after the intelligent customer service model is built, a response test can be performed on the intelligent customer service model to test whether a response result provided by the intelligent customer service model meets expectations, if yes, the intelligent customer service model is determined to pass the response test, the intelligent customer service model is stored in a designated storage area so that a server loads the intelligent customer service model from the designated storage area, and message processing is performed on a message to be processed through the intelligent customer service model to obtain a response result corresponding to the message to be processed. Wherein the designated storage area may be a database or a local storage area.

For the understanding of the foregoing embodiments, the embodiment of the present invention provides an application example of a method for constructing an intelligent customer service model, referring to a flowchart of another method for constructing an intelligent customer service model shown in fig. 2, the method mainly includes steps S202 to S212:

Step S202, selecting a flow node type to generate a flow node. The flow node type, namely the type of the candidate functional component, comprises one or more of a message sending component, a problem collecting component, an interface calling component, a branch jump component, a display component and a code customization component. Specifically, the messaging component, i.e., sendMessage component, may send rich messages to clients, such as clients that support plain text and picture format messages, or clients that support rich text and picture format messages (e.g., web (World Wide Web) ends); the problem collection component is a collection Input component and is used for collecting client information through specified problems, supporting various verification modes such as dates, mailboxes, regular expressions and the like, and allowing clients to retry Input; the interface calling component, namely the serviceall component, can send an HTTP request to call API interfaces of other systems, support variable parameters and extract appointed information in HTTP response data through JSON PATH to store the appointed information in the session context; the Branch jump component is used for Branch jump, sets one or more conditions (complex judgment is carried out according to the variables in the session context) and determines the subsequent customer service support flow; the display component, namely the JSON Output component, can be formatted into message Output aiming at HTTP JSON response, and supports template rendering and cyclic rendering; the code custom component, namely the Python Script component, can support a user to inject custom codes in a complex scene. Through use verification, most business scenarios can be satisfied based on the 6 different types of low-code components (flow nodes).

In step S204, two different flow nodes are associated. In one embodiment, a flow relationship between flow nodes may be generated by associating two different flow nodes.

Step S206, editing attribute parameter values of the flow nodes. In one embodiment, the attribute items owned by different types of flow nodes are also different. For example, referring to a dialog flow diagram of an intelligent customer service model shown in fig. 3, the intelligent customer service model includes the following procedures: inquiring weather, collecting appointed city information, judging whether to inquire quickly, inquiring Guangdong river area/inquiring appointed area, and outputting weather information.

Step S208, testing the dialogue flow of the intelligent customer service model. In one embodiment, the dialog flow of the intelligent customer service model can be verified to be correct by simulating the dialog. For example, referring to a test interface schematic diagram of an intelligent customer service model shown in fig. 4, when a keyword "@ test weather" is input, a reply "please input a city code to be queried, support quick query [ XXX ], and when a keyword" XXX "is input, reply" the queried weather information is as follows: province Guangdong, urban Tianhe area, weather, temperature 29 and humidity 79%, thereby judging that the dialogue flow of the intelligent customer service model is correct. The embodiment of the invention also provides an online dialogue effect schematic diagram of the intelligent customer service model shown in fig. 5, and the online dialogue effect can also display information such as message sending time, user identification, user head portrait, read/unread and the like.

Step S210, the intelligent customer service model is stored in a database.

And S212, releasing the intelligent customer service model, analyzing and loading the intelligent customer service model in the database to the message processing module, and waiting for message triggering.

In an alternative embodiment, the dialogue record may be displayed through a graphical user interface, so as to facilitate tracing the tracing problem, for example, referring to a record query interface schematic diagram of an intelligent customer service model shown in fig. 6, in a scene of querying the dialogue record, specific details of the dialogue record, such as saving variables, initiating requests, service responses, and the like, may be displayed through the graphical user interface, and a final response result may be displayed, so that when the response result is inaccurate, the problem existing in the intelligent customer service model may be accurately located.

In one embodiment, referring to a method for constructing an intelligent customer service model shown in fig. 7, the architecture is divided into two nodes: a build phase (i.e., the build method of the intelligent customer service model described above) and a release run phase (i.e., the message processing method). FIG. 7 illustrates a method for constructing an intelligent customer service model according to an embodiment of the present invention, in which engineers perform abstract analysis on a business process to formulate a dialogue interaction rule between a customer service robot and a user; then, a trigger flow is constructed through a robot rule arrangement module, and the construction of the flow can be completed through the drag arrangement through triggering keywords (namely, keywords trigger), sending messages to users (namely, asking the users for send msg), throwing out problem to collect user input (namely, collecting user input collection), branching jump (namely, branching jump branch) to low code components such as different sub-flows (comprising branch1 and branch 2) and the like; and finally, converting the data format of the flow node, and storing the flow node information and the connection side information into a MongoDB database in a JSON format for loading rules when the robot runs.

In addition, considering that the existing message processing method is mainly provided by operators on duty for manual support, the following defects or demand scenes mainly exist: (1) The operator on duty cannot provide real-time support in non-working time, and the operator on duty can face the embarrassment of delayed response aiming at some sudden business problems, so that the overall service quality is difficult to guarantee; (2) As departments become increasingly involved in undertaking business and new staff join, related staff spend a great deal of time each day for answering basic questions, and besides the relatively labor cost, the uncertainty of manual processing cannot guarantee the quality of service; (3) When providing service support, the dialogue between the attendant and the business personnel is not uniformly recorded and analyzed, which means that the business problems can not be summarized through historical data, the support mode is optimized, and the service effect is improved; (4) Although the existing system supports the man-hour recording of each service by the attendant, the manual filling is needed, the mode is not friendly to the attendant, and the man-hour of the manual filling has a certain error with the real data.

In order to improve the above problem, the embodiment of the present invention further provides a message processing method, which is applied to a server, referring to a flow chart of a message processing method shown in fig. 8, the method mainly includes the following steps S802 to S804:

Step S802, at least one message to be processed is monitored. In one embodiment, the message to be processed may be monitored by the next message monitoring module.

Step S804, message processing is carried out on each message to be processed through a pre-constructed intelligent customer service model, and a response result corresponding to each message to be processed is obtained. The intelligent customer service model is obtained by loading a server from a designated storage area, and is obtained by constructing by adopting a construction method of the intelligent customer service model. In one embodiment, the message processing module may use the intelligent customer service model to process the messages to be processed in parallel, and in specific implementation, the message processing module may trigger the corresponding intelligent customer service model according to the keywords in the messages to be processed, optionally, the intelligent customer service models corresponding to different keywords are different, and then the message processing module uses the corresponding intelligent customer service model to process the messages to be processed respectively, so as to obtain the response result.

Compared with the prior art that the service quality cannot be guaranteed by manual processing, the message processing method provided by the embodiment of the invention can remarkably improve the service effect and remarkably reduce the labor cost consumed by message processing.

With continued reference to fig. 7, fig. 7 also illustrates that the server is configured with a message listening module, a message adapter, a message processing module, a load balancer, and a monitoring module. In a specific implementation, the message processing module (1) loads an intelligent customer service model (also called as a dialogue flow rule) of the robot from the mongo db; (2) The message monitoring module monitors messages from chat media such as clients and the like through different communication protocols; (3) The message adapter module processes the user messages from different media and converts the user messages into a unified message format; (4) Forwarding the message to a message processing module for concurrent processing through a load balancer (load balancer); (5) The message processing module executes different tasks and flows the state through the loaded rule and the current session state of the user; (6) The message processing module can upload index data generated in the processing process to the monitoring module for recording, and the monitoring module can send out a warning to a robot manager when the threshold value is reached.

In order to facilitate understanding of step S802, an embodiment of the present invention provides an implementation manner of monitoring at least one message to be processed, where the message monitoring module may monitor at least one message to be processed sent by a target terminal based on a communication protocol corresponding to the target terminal, and forward each message to be processed to a message adapter. Specifically, referring to a message listening module architecture diagram shown in fig. 9, the message listening module is responsible for capturing user messages from different chat mediums and forwarding the captured messages to downstream module processing (i.e., message adapters). Optionally, a communication protocol may be used to simulate a login event through Socket, after the related server side authentication KEY is obtained, the subsequent sending messages are decrypted and encrypted through the KEY, and the user identity, the single session message, the group invitation and the like can be identified according to the message format. Optionally, for the Web end, only the user authenticated by the Auth system can trigger the customer service robot, that is, the message monitoring module can identify the user identity based on the user authentication, the user establishes WebSocket connection with the message monitoring module when initiating a session to the customer service robot, and subsequent message receiving and sending can be performed through the connection.

In one embodiment, before executing the step of processing the message of each message to be processed through the pre-built intelligent customer service model, format conversion can be performed on each message to be processed through the message adapter based on the communication protocol, so as to obtain the message to be processed in the specified format. In practical application, the message adapter module performs unified format processing on messages of different chat mediums, extracts corresponding values (such as user ID (Identity document, identity number), chat channel, etc.) from a communication protocol used by the chat mediums, saves the values in the structured data, and then inputs the messages in the unified format into the message processing module.

In another embodiment, after the step of performing message processing on each message to be processed through the pre-built intelligent customer service model, format conversion can be performed on the response result through the message adapter based on the communication protocol, so as to obtain a response result which can be identified by the target terminal. In practical application, the message replied by the message processing module is also converted by the message adapter into a format conforming to a different chat medium communication protocol and sent to the user. In the embodiment of the invention, the message adapter module adopts an adapter mode in a design mode, has good expansibility and can be accessed into a new chat medium at any time.

In order to facilitate understanding the foregoing step S804, the embodiment of the present invention further provides an implementation manner of performing message processing on each message to be processed through a pre-built intelligent customer service model to obtain a response result corresponding to each message to be processed, which is referred to in steps 1 to 3 below:

and step 1, forwarding the message to be processed in the specified format to a load balancer through a message adapter.

And 2, forwarding at least one target message to be processed in the specified format to the message processing module through a load balancer according to the current load of the message processing module. For example, when the current load of the message processor is large, no pending messages or a small number of pending messages may be allocated. The embodiment of the invention can avoid the overload of the message processor to a certain extent by configuring the load equalizer, thereby ensuring the message processor to stably operate.

And 3, performing message processing on each target message to be processed based on the intelligent customer service model through a message processing module to obtain a response result corresponding to each target message to be processed. The message processing module stores a pre-built intelligent customer service model in a state circulation table, and the user states in the state circulation table are in one-to-one correspondence with flow nodes in the intelligent customer service model. In an alternative embodiment, reference may be made to the following steps 3.1 to 3.4:

And 3.1, determining the next user state to be circulated from the state circulation table based on the target to-be-processed message and the current user state of the target to-be-processed message, and operating a flow node corresponding to the next user state. For ease of understanding, reference is made to a message handling module architecture diagram shown in fig. 10, in which it is responsible for handling unified format messages entered by a message adapter module. When the robot is initialized, the flow rule is loaded from the database and stored in the memory in the form of a state flow table, and then the robot is in a waiting state. When the message adapter module sends the message in the unified format to the message processing module, the current state of the user to which the message belongs is read from the Redis cache, the message input by the user is used as a condition, and the message is transferred to the next state through the finite state machine according to the current state and the condition.

And 3.2, under the condition that the operation of the flow node corresponding to the next user state is completed and the user input event is triggered, acquiring the feedback information input by the user, and continuously determining the next user state to be circulated from the state circulation table based on the feedback information. In one embodiment, each user state corresponds to a process node (sending a message, calling an interface, etc.) in an intelligent customer service model, after the task corresponding to the process node is executed, whether the next state needs user input is judged, if so, the current state is written back into the Redis cache, the next message (i.e. feedback message) is waited, and then state circulation is performed according to the next message. It should be noted that the feedback message also needs to be subjected to format conversion by the message adapter and then subjected to message processing by the message processing module.

And 3.3, under the condition that the operation of the flow node corresponding to the next user state is completed and the user input event is not triggered, continuously determining the next user state to be circulated from the state circulation table based on the target message to be processed. In one embodiment, if no input is required, the message entered by the user is conditioned to continue to flow through the finite state machine to the next state according to the current state and condition.

And 3.4, obtaining a response result corresponding to the target message to be processed until a preset circulation ending condition is met. The circulation ending condition can be that the task corresponding to the last flow node is operated and completed.

In one embodiment, during the message processing process, the monitoring module may monitor whether the index data generated during the message processing process reaches a preset threshold, and when the judgment result is yes, generate a message processing alarm prompt, and send the message processing alarm prompt to the designated associated terminal. The index data may be the number of times of correctly processing the user message, the number of times of incorrectly processing the user message, the number of times of correctly processing the user message and the number of times of errors occurring in 1 minute, 5 minutes and 10 minutes, and the like, and may be specifically configured based on actual requirements. By way of example, assuming that the number of errors exceeds 10 times within 1 minute, an alarm may be issued by mail or the like the administrator of the intelligent customer service model.

For ease of understanding, an application example of message processing is provided in the embodiment of the present invention, and referring to a schematic diagram of an intelligent customer service model shown in fig. 11, each intelligent customer service model may include a plurality of flow nodes, where each flow node is generated by dragging a low code component, and a circulation relationship exists between two flow nodes. Taking the intelligent customer service model corresponding to the server hardware details in fig. 11 as an example: (1) Referring to the schematic diagram of another intelligent customer service model shown in fig. 12, the triggering keyword of the intelligent customer service model corresponding to the server hardware details is @ srv_hw, and when the message to be processed contains "@ srv_hw" word, the intelligent customer service model corresponding to the triggering server hardware details is determined. (2) Referring to the schematic diagram of another intelligent customer service model shown in fig. 13, a flow node "Athena interface INFO" in the intelligent customer service model will call a third party system to query the hardware information of the server, and a corresponding interface address needs to be filled in the attribute parameter value of the flow node "Athena interface INFO" to call the corresponding third party system by using the interface address, and save the interface response. (3) Referring to the schematic diagram of another intelligent customer service model shown in fig. 14, a branch determination is performed by identifying parameters, and by way of example, if the parameters include all, the process jumps to the all branch module, and if the parameters include product, the process jumps to the product branch module. (4) Referring to another intelligent customer service module shown in fig. 15, taking a jump to all branch module as an example, the all branch module formats and replies server hardware parameters in response to an output format, such as a response interface of an intelligent customer service model shown in fig. 16, where the queried server hardware parameters are displayed.

In summary, the embodiment of the invention provides a flexible and light chat customer service robot construction scheme for business personnel or developers with zero development experience, and the method comprises the full life cycle management of chat robot arrangement rules, model training, testing, online release and log record. In concrete implementation, the intelligent customer service generation scheme based on the Low-Code visualization component is an independently developed customer service robot construction framework, and the framework consists of a customer service robot rule arrangement module, a message monitoring module, a message adapter and a message processing module, and is combined with an interactive interface in a Low-Code mode to support a user to finish construction, management, test and release of a brand-new customer service robot within 15 minutes. The method for constructing the intelligent customer service model and the method for processing the message provided by the embodiment of the invention have the following characteristics:

(1) Low code: supporting user drag definition of customer service dialogue rules and step flows under different scenes without any research and development experience according to various front-end interaction components provided by the framework;

(2) The method comprises the following steps of: aiming at complex business logic, a developer user is allowed to realize the method by injecting codes or calling an API;

(3) And (5) hot updating: in addition to supporting minute-level construction and deployment, aiming at updating of customer service robot rules, second-level non-stop thermal updating is supported, and iterative updating of the customer service robot is responded quickly;

(4) Intelligent: based on the Turing API, the intelligent reply can be made to daily dialogue, the user is supported to upload the corpus to train the robot model, the robot is endowed with vivid individuality, and the customer dialogue hit rate is improved.

For the method for constructing the intelligent customer service model provided in the foregoing embodiment, the embodiment of the present invention provides a device for constructing the intelligent customer service model, and a graphical user interface is provided through a terminal, and referring to a schematic structural diagram of the device for constructing the intelligent customer service model shown in fig. 17, the device mainly includes the following parts:

a component display module 1702 for displaying a plurality of candidate functional components via a graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model;

a node generation module 1704 for determining at least one target functional component in response to a drag operation for the candidate functional component and generating a flow node based on the target functional component;

the association editing module 1706 is configured to establish an association relationship between flow nodes in response to an association operation for the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations for the flow nodes;

The model building module 1708 is configured to build an intelligent customer service model based on the association relationship between the flow nodes and the attribute parameter values.

The intelligent customer service model constructing device provided by the embodiment of the invention constructs the intelligent customer service model in a dragging and arranging mode of the functional components with rich types, can rapidly meet service scenes, can effectively reduce development difficulty of the intelligent customer service model, reduces development and updating periods of the intelligent customer service model, and effectively reduces labor cost and time cost required by constructing the intelligent customer service model.

In one embodiment, the candidate functional components include one or more of a messaging component, a problem collection component, an interface invocation component, a branch jump component, a display component, a code customization component.

In one embodiment, the apparatus further includes a test module configured to: performing response test on the intelligent customer service model; if the intelligent customer service model passes the response test, the intelligent customer service model is stored in a designated storage area so that the server can load the intelligent customer service model from the designated storage area, and the message processing is carried out on the message to be processed through the intelligent customer service model, so that a response result corresponding to the message to be processed is obtained.

For the message processing method provided in the foregoing embodiment, the embodiment of the present invention provides a message processing apparatus applied to a server, referring to a schematic structural diagram of a message processing apparatus shown in fig. 18, the apparatus mainly includes the following parts:

a message listening module 1802 configured to listen to at least one message to be processed;

the message processing module 1804 is configured to process a message for each message to be processed through a pre-built intelligent customer service model, so as to obtain a response result corresponding to each message to be processed; the intelligent customer service model is obtained by loading a server from a designated storage area, and is constructed by adopting the construction method of the intelligent customer service model in any one of claims 1-3.

The message processing device provided by the embodiment of the invention responds to the message to be processed by utilizing the intelligent customer service model, and compared with the problem that the service quality cannot be ensured in the manual processing in the prior art, the message processing device provided by the embodiment of the invention can remarkably improve the service effect and remarkably reduce the labor cost consumed by message processing.

In one embodiment, the message listening module 1802 is further configured to listen to at least one pending message sent by the target terminal based on a communication protocol corresponding to the target terminal, and forward each pending message to the message adapter.

In one embodiment, the message adapter is further configured to perform format conversion on each message to be processed based on the communication protocol, to obtain a message to be processed in a specified format; the message adapter is further used for performing format conversion on the response result based on the communication protocol to obtain a response result which can be identified by the target terminal.

In one embodiment, the message adapter is further configured to forward the message to be processed in the specified format to the load balancer; the load balancer is further used for forwarding at least one target message to be processed in the message to be processed in a specified format to the message processing module according to the current load of the message processing module; the message processing module is further used for processing the message of each target message to be processed based on the intelligent customer service model to obtain a response result corresponding to each target message to be processed.

In one implementation mode, a message processing module stores a pre-constructed intelligent customer service model in a form of a state circulation table, and user states in the state circulation table are in one-to-one correspondence with flow nodes in the intelligent customer service model; the message processing module is further configured to: determining the next user state to be circulated from a state circulation table based on the target to-be-processed message and the current user state of the target to-be-processed message, and operating a flow node corresponding to the next user state; when the operation of the flow node corresponding to the next user state is completed and a user input event is triggered, acquiring a feedback message input by the user, and continuously determining the next user state to be circulated from a state circulation table based on the feedback message; under the condition that the operation of the flow node corresponding to the next user state is completed and the user input event is not triggered, continuously determining the next user state to be circulated from the state circulation table based on the target message to be processed; and obtaining a response result corresponding to the target message to be processed until a preset circulation ending condition is met.

In one embodiment, the monitoring module is further configured to: monitoring whether index data generated in the message processing process reaches a preset threshold value or not in the message processing process; if so, generating a message processing alarm prompt and sending the message processing alarm prompt to the appointed associated terminal.

The device provided by the embodiment of the present invention has the same implementation principle and technical effects as those of the foregoing method embodiment, and for the sake of brevity, reference may be made to the corresponding content in the foregoing method embodiment where the device embodiment is not mentioned.

The embodiment of the invention provides electronic equipment, which comprises a processor and a storage device; the storage means has stored thereon a computer program which, when executed by the processor, performs:

a construction method of an intelligent customer service model provides a graphical user interface through a terminal, comprising the following steps: displaying a plurality of candidate functional components through a graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model; determining at least one target functional component in response to a drag operation for the candidate functional component, and generating a flow node based on the target functional component; responding to the association operation aiming at the flow nodes, and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations for the flow nodes; and constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values.

In one embodiment, the method further comprises: performing response test on the intelligent customer service model; if the intelligent customer service model passes the response test, the intelligent customer service model is stored in a designated storage area so that the server can load the intelligent customer service model from the designated storage area, and the message processing is carried out on the message to be processed through the intelligent customer service model, so that a response result corresponding to the message to be processed is obtained.

According to the electronic equipment provided by the embodiment of the invention, the intelligent customer service model is constructed in a dragging and arranging mode of the functional components with rich types, so that the service scene can be rapidly met, the development difficulty of the intelligent customer service model can be effectively reduced, the development and updating period of the intelligent customer service model can be reduced, and the labor cost and the time cost required by constructing the intelligent customer service model can be effectively reduced.

A message processing method, the method being applied to a server, the method comprising: monitoring at least one message to be processed; message processing is carried out on each message to be processed through a pre-built intelligent customer service model, and a response result corresponding to each message to be processed is obtained; the intelligent customer service model is obtained by loading a server from a designated storage area, and is constructed by adopting the construction method of any intelligent customer service model provided in the first aspect.

In one embodiment, a server is configured with a message listening module and a message adapter; a step of listening to at least one message to be processed, comprising: and monitoring at least one message to be processed sent by the target terminal through the message monitoring module based on a communication protocol corresponding to the target terminal, and forwarding each message to be processed to the message adapter.

In one embodiment, before the step of processing each message to be processed by the pre-built intelligent customer service model, the method further comprises: converting the format of each message to be processed based on a communication protocol through a message adapter to obtain the message to be processed in a specified format; after the step of processing each message to be processed by the pre-built intelligent customer service model, the method further comprises the following steps: and converting the format of the response result based on the communication protocol through the message adapter to obtain the response result which can be identified by the target terminal.

In one embodiment, the server is further configured with a message processing module and a load balancer; the method comprises the steps of carrying out message processing on each message to be processed through a pre-constructed intelligent customer service model to obtain a response result corresponding to each message to be processed, and the method comprises the following steps: forwarding a message to be processed in a specified format to a load balancer through a message adapter; forwarding at least one target message to be processed in the message to be processed in a specified format to the message processing module through a load balancer according to the current load of the message processing module; and carrying out message processing on each target message to be processed based on the intelligent customer service model through a message processing module to obtain a response result corresponding to each target message to be processed.

In one implementation mode, a message processing module stores a pre-constructed intelligent customer service model in a form of a state circulation table, and user states in the state circulation table are in one-to-one correspondence with flow nodes in the intelligent customer service model; message processing is carried out on each target message to be processed based on the intelligent customer service model, and a response result corresponding to each target message to be processed is obtained, which comprises the following steps: determining the next user state to be circulated from a state circulation table based on the target to-be-processed message and the current user state of the target to-be-processed message, and operating a flow node corresponding to the next user state; when the operation of the flow node corresponding to the next user state is completed and a user input event is triggered, acquiring a feedback message input by the user, and continuously determining the next user state to be circulated from a state circulation table based on the feedback message; under the condition that the operation of the flow node corresponding to the next user state is completed and the user input event is not triggered, continuously determining the next user state to be circulated from the state circulation table based on the target message to be processed; and obtaining a response result corresponding to the target message to be processed until a preset circulation ending condition is met.

In one embodiment, the server is further configured with a monitoring module; the method further comprises the steps of: monitoring whether index data generated in the message processing process reach a preset threshold value or not through a monitoring module in the message processing process; if so, generating a message processing alarm prompt and sending the message processing alarm prompt to the appointed associated terminal.

Compared with the prior art that the service quality cannot be guaranteed by manual processing, the electronic equipment provided by the embodiment of the invention can remarkably improve the service effect and remarkably reduce the labor cost consumed by message processing.

Fig. 19 is a schematic structural diagram of an electronic device according to an embodiment of the present invention, where the electronic device 100 includes: a processor 190, a memory 191, a bus 192 and a communication interface 193, the processor 190, the communication interface 193 and the memory 191 being connected by the bus 192; processor 190 is configured to execute executable modules, such as computer programs, stored in memory 191.

The memory 191 may comprise a high-speed random access memory (RAM, random Access Memory), and may further comprise a non-volatile memory (non-volatile memory), such as at least one magnetic disk memory. The communication connection between the system network element and at least one other network element is implemented via at least one communication interface 193 (which may be wired or wireless), which may use the internet, a wide area network, a local network, a metropolitan area network, etc.

Bus 192 may be an ISA bus, a PCI bus, an EISA bus, or the like. The buses may be classified as address buses, data buses, control buses, etc. For ease of illustration, only one bi-directional arrow is shown in fig. 19, but not only one bus or one type of bus.

The memory 191 is configured to store a program, and the processor 190 executes the program after receiving an execution instruction, and the method executed by the apparatus for flow defining disclosed in any of the foregoing embodiments of the present invention may be applied to the processor 190 or implemented by the processor 190.

Processor 190 may be an integrated circuit chip having signal processing capabilities. In implementation, the steps of the methods described above may be performed by integrated logic circuitry in hardware in processor 190 or by instructions in software. The processor 190 may be a general-purpose processor, including a central processing unit (Central Processing Unit, CPU), a network processor (Network Processor, NP), etc.; but may also be a digital signal processor (Digital Signal Processing, DSP for short), application specific integrated circuit (Application Specific Integrated Circuit, ASIC for short), off-the-shelf programmable gate array (Field-Programmable Gate Array, FPGA for short), or other programmable logic device, discrete gate or transistor logic device, discrete hardware components. The disclosed methods, steps, and logic blocks in the embodiments of the present invention may be implemented or performed. A general purpose processor may be a microprocessor or the processor may be any conventional processor or the like. The steps of the method disclosed in connection with the embodiments of the present invention may be embodied directly in the execution of a hardware decoding processor, or in the execution of a combination of hardware and software modules in a decoding processor. The software modules may be located in a random access memory, flash memory, read only memory, programmable read only memory, or electrically erasable programmable memory, registers, etc. as well known in the art. Which is located in memory 191 and processor 190 reads information from memory 191 to perform the steps of the method described above in connection with its hardware.

A computer program product of a readable storage medium according to an embodiment of the present invention includes a computer readable storage medium storing program code including instructions operable to perform:

The readable storage medium provided by the embodiment of the invention constructs the intelligent customer service model in a dragging arrangement mode of the functional components with rich types, can rapidly meet service scenes, can effectively reduce development difficulty of the intelligent customer service model, reduce development and update periods of the intelligent customer service model, and effectively reduce labor cost and time cost required by constructing the intelligent customer service model.

The readable storage medium provided by the embodiment of the invention responds to the message to be processed by utilizing the intelligent customer service model, and compared with the problem that the service quality cannot be ensured in the manual processing in the prior art, the embodiment of the invention can remarkably improve the service effect and remarkably reduce the labor cost consumed by message processing.

The functions, if implemented in the form of software functional units and sold or used as a stand-alone product, may be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention may be embodied essentially or in a part contributing to the prior art or in a part of the technical solution, in the form of a software product stored in a storage medium, comprising several instructions for causing a computer device (which may be a personal computer, a server, a network device, etc.) to perform all or part of the steps of the method according to the embodiments of the present invention. And the aforementioned storage medium includes: a U-disk, a removable hard disk, a Read-Only Memory (ROM), a random access Memory (RAM, random Access Memory), a magnetic disk, or an optical disk, or other various media capable of storing program codes.

Finally, it should be noted that: the above examples are only specific embodiments of the present invention, and are not intended to limit the scope of the present invention, but it should be understood by those skilled in the art that the present invention is not limited thereto, and that the present invention is described in detail with reference to the foregoing examples: any person skilled in the art may modify or easily conceive of the technical solution described in the foregoing embodiments, or perform equivalent substitution of some of the technical features, while remaining within the technical scope of the present disclosure; such modifications, changes or substitutions do not depart from the spirit and scope of the technical solutions of the embodiments of the present invention, and are intended to be included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the claims.

Claims

1. The method for constructing the intelligent customer service model is characterized by providing a graphical user interface through a terminal, and comprising the following steps:

displaying a plurality of candidate functional components through the graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model;

Determining at least one target functional component in response to a drag operation for the candidate functional component, and generating a flow node based on the target functional component;

responding to the association operation aiming at the flow nodes, and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations directed to the flow nodes;

and constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values.

2. The method of claim 1, wherein the candidate functionality component comprises one or more of a messaging component, a problem collection component, an interface invocation component, a branch jump component, a display component, a code customization component.

3. The method according to claim 1, wherein the method further comprises:

performing response test on the intelligent customer service model;

if the intelligent customer service model passes the response test, the intelligent customer service model is stored in a designated storage area so that a server can load the intelligent customer service model from the designated storage area, and the message processing is carried out on the message to be processed through the intelligent customer service model, so that a response result corresponding to the message to be processed is obtained.

4. A message processing method, wherein the method is applied to a server, the method comprising:

monitoring at least one message to be processed;

message processing is carried out on each message to be processed through a pre-built intelligent customer service model, and a response result corresponding to each message to be processed is obtained; the intelligent customer service model is obtained by loading the server from a designated storage area, and is constructed by adopting the construction method of the intelligent customer service model in any one of claims 1-3.

5. The method of claim 4, wherein the server is configured with a message listening module and a message adapter; the step of listening for at least one message to be processed comprises:

and monitoring at least one message to be processed sent by the target terminal through the message monitoring module based on a communication protocol corresponding to the target terminal, and forwarding each message to be processed to the message adapter.

6. The method of claim 5, wherein prior to the step of message processing each of the pending messages by a pre-built intelligent customer service model, the method further comprises:

Converting the format of each message to be processed based on the communication protocol through the message adapter to obtain the message to be processed in a specified format;

after the step of processing each message to be processed by the pre-built intelligent customer service model, the method further comprises the following steps:

and converting the format of the response result based on the communication protocol through the message adapter to obtain a response result which can be identified by the target terminal.

7. The method of claim 6, wherein the server is further configured with a message processing module and a load balancer; the step of processing the message of each message to be processed through the pre-constructed intelligent customer service model to obtain a response result corresponding to each message to be processed comprises the following steps:

forwarding the message to be processed in the specified format to the load balancer through the message adapter;

forwarding, by the load balancer, at least one target pending message of the pending messages in the specified format to the message processing module according to a current load of the message processing module;

And carrying out message processing on each target message to be processed based on the intelligent customer service model through the message processing module to obtain a response result corresponding to each target message to be processed.

8. The method of claim 7, wherein the message processing module stores a pre-built intelligent customer service model in the form of a state flow table, and user states in the state flow table are in one-to-one correspondence with flow nodes in the intelligent customer service model; the step of processing the message of each target message to be processed based on the intelligent customer service model to obtain a response result corresponding to each target message to be processed comprises the following steps:

determining a next user state to be circulated from the state circulation table based on the target to-be-processed message and the current user state of the target to-be-processed message, and operating the flow node corresponding to the next user state;

when the operation of the flow node corresponding to the next user state is completed and a user input event is triggered, acquiring a feedback message input by a user, and continuously determining the next user state to be circulated from the state circulation table based on the feedback message;

When the operation of the flow node corresponding to the next user state is completed and a user input event is not triggered, continuously determining the next user state to be circulated from the state circulation table based on the target message to be processed;

and obtaining a response result corresponding to the target message to be processed until a preset circulation ending condition is met.

9. The method of claim 4, wherein the server is further configured with a monitoring module; the method further comprises the steps of:

monitoring whether index data generated in the message processing process reach a preset threshold value or not through the monitoring module in the message processing process;

if yes, generating a message processing alarm prompt, and sending the message processing alarm prompt to a specified associated terminal.

10. The device for constructing the intelligent customer service model is characterized by providing a graphical user interface through a terminal, and comprising the following steps:

a component display module for displaying a plurality of candidate functional components through the graphical user interface; the candidate function component is used for providing functions required for constructing the intelligent customer service model;

the node generation module is used for responding to the drag operation aiming at the candidate functional components, determining at least one target functional component and generating a flow node based on the target functional component;

The association editing module is used for responding to association operation aiming at the flow nodes and establishing association relation among the flow nodes; and determining attribute parameter values for the flow nodes in response to editing operations directed to the flow nodes;

and the model construction module is used for constructing an intelligent customer service model based on the association relation between the flow nodes and the attribute parameter values.

11. A message processing apparatus, the apparatus being applied to a server, the apparatus comprising:

the message monitoring module is used for monitoring at least one message to be processed;

the message processing module is used for processing the message of each message to be processed through a pre-constructed intelligent customer service model to obtain a response result corresponding to each message to be processed; the intelligent customer service model is obtained by loading the server from a designated storage area, and is constructed by adopting the construction method of the intelligent customer service model in any one of claims 1-3.

12. An electronic device comprising a processor and a memory, the memory storing computer-executable instructions executable by the processor, the processor executing the computer-executable instructions to implement the method of any one of claims 1 to 3, 4 to 9.

13. A computer readable storage medium storing computer executable instructions which, when invoked and executed by a processor, cause the processor to implement the method of any one of claims 1 to 3, 4 to 9.