Business question and answer processing method, system and equipment and storage medium

Abstract

The embodiment of the invention discloses a business question and answer processing method, system and device and a storage medium. According to the technical scheme provided by the embodiment of the invention, the target problem vector value of the target service problem is determined by receiving the target service problem, the pre-constructed question and answer database and the function knowledge base are searched based on the target problem vector value, and the question and answer database configures reply data corresponding to different problem vectors in advance; the function knowledge base configures business functions corresponding to different problem vectors in advance; determining first reply data matched with the target question vector value from a question and answer database, and / or determining target function information matched with the target question vector value from a function knowledge base, executing a specified service based on the target function information, and determining second reply data; and replying the service problem based on the first reply data and / or the second reply data. By adopting the technical means, the service processing function of the intelligent dialogue assistant can be expanded, the service processing effect is improved, and the service question and answer experience of the user is improved.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of computer technology, and in particular, to a business question and answer processing method, system, device and storage medium. Background Art

[0002] At present, with the development of various business functions of applications, more and more business problems need to be queried and processed in the process of business operation and management, such as querying user ban history, querying user reward records or bonus records, etc. When handling various business problems, the intelligent dialogue assistant is used to conduct business Q&A. According to the business questions raised by the user, the intelligent dialogue assistant matches the business questions according to the question and answer knowledge base, and then returns the corresponding answer results to the user.

[0003] However, processing business problems based solely on a question-and-answer knowledge base is difficult to support the query and operation functions of complex businesses. Its processing functions are relatively simple, the business processing results are biased, and it is difficult to meet users' complex business processing needs. Summary of the invention

[0004] The embodiments of the present application provide a business question and answer processing method, system, device and storage medium, which can query and operate complex business, and solve the technical problem that intelligent dialogue assistants have difficulty supporting the processing of complex business queries and operations.

[0005] In a first aspect, an embodiment of the present application provides a business question and answer processing method, including:

[0006] Receive a target business question, determine a target question vector value of the target business question, search a pre-built question-answering database and a function knowledge base based on the target question vector value, wherein the question-answering database pre-configures reply data corresponding to different question vectors, and the function knowledge base pre-configures business functions corresponding to different question vectors;

[0007] Determine first reply data matching the target question vector value from the question-answer database, and / or determine target function information matching the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data;

[0008] The business question is responded to based on the first response data and / or the second response data.

[0009] In a second aspect, an embodiment of the present application provides a business question and answer processing system, including:

[0010] A search module is configured to receive a target business question, determine a target question vector value of the target business question, and search a pre-built question-answer database and a function knowledge base based on the target question vector value, wherein the question-answer database is pre-configured with response data corresponding to different question vectors, and the function knowledge base is pre-configured with business functions corresponding to different question vectors;

[0011] a matching module configured to determine first reply data matching the target question vector value from the question-answer database, and / or determine target function information matching the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data;

[0012] The reply module is configured to reply to the business question based on the first reply data and / or the second reply data.

[0013] In a third aspect, an embodiment of the present application provides a business question and answer processing device, including:

[0014] memory and one or more processors;

[0015] The memory is configured to store one or more programs;

[0016] When the one or more programs are executed by the one or more processors, the one or more processors implement the business question and answer processing method as described in the first aspect.

[0017] In a fourth aspect, an embodiment of the present application provides a non-volatile computer-readable storage medium, which stores computer-executable instructions, and the computer-executable instructions, when executed by a computer processor, are configured to execute the business question and answer processing method as described in the first aspect.

[0018] In a fifth aspect, an embodiment of the present application provides a computer program product, which includes instructions. When the instructions are executed on a computer or a processor, the computer or the processor executes the business question and answer processing method as described in the first aspect.

[0019] The embodiment of the present application receives a target business question, determines a target question vector value of the target business question, searches a pre-constructed question-and-answer database and a function knowledge base based on the target question vector value, wherein the question-and-answer database pre-configures reply data corresponding to different question vectors, and the function knowledge base pre-configures business functions corresponding to different question vectors; determines first reply data that matches the target question vector value from the question-and-answer database, and / or determines target function information that matches the target question vector value from the function knowledge base, executes a designated business based on the target function information, and determines second reply data; and replies to the business question based on the first reply data and / or the second reply data. By adopting the above-mentioned technical means, the designated business is executed in combination with the corresponding business function, and the reply data of the business question is determined to reply to the business question, so as to expand the business processing function of the intelligent dialogue assistant, realize the query and operation of complex business and other functions through the corresponding business function, improve the business processing effect, and enhance the user's business question-and-answer experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] Figure 1 It is a flow chart of a business question and answer processing method provided by an embodiment of the present application;

[0021] Figure 2 It is a schematic diagram of constructing a question-and-answer database in an embodiment of the present application;

[0022] Figure 3 It is a schematic diagram of constructing a function knowledge base in an embodiment of the present application;

[0023] Figure 4 This is a business question and answer processing flow chart of the intelligent dialogue assistant in the embodiment of the present application;

[0024] Figure 5 It is a structural diagram of a business question and answer processing system provided in an embodiment of the present application;

[0025] Figure 6 It is a structural diagram of a business question and answer processing device provided in an embodiment of the present application. DETAILED DESCRIPTION

[0026] In order to make the purpose, technical scheme and advantages of the present application clearer, the specific embodiments of the present application are further described in detail below in conjunction with the accompanying drawings. It is understood that the specific embodiments described herein are only used to explain the present application, rather than to limit the present application. It should also be noted that, for the convenience of description, only the part related to the present application but not all the contents are shown in the accompanying drawings. Before discussing the exemplary embodiments in more detail, it should be mentioned that some exemplary embodiments are described as processes or methods depicted as flow charts. Although the flow chart describes each operation (or step) as a sequential process, many of the operations therein can be implemented in parallel, concurrently or simultaneously. In addition, the order of each operation can be rearranged. The process can be terminated when its operation is completed, but it can also have additional steps not included in the accompanying drawings. The process can correspond to a method, a function, a procedure, a subroutine, a subprogram, etc.

[0027] The business question and answer processing method provided in this application aims to execute specified business by combining corresponding business functions, determine the response data of business questions to answer business questions, thereby expanding the business processing functions of the intelligent dialogue assistant and realizing functions such as query and operation of complex business through corresponding business functions.

[0028] As the business functions of applications develop, there are many business problems that require developers to assist in querying in the operation and management of short-term and long-term activities of applications. For example, querying the user's ban history, querying the user's reward record or bonus record, etc. These problems occur frequently and are repetitive, but the query method is relatively complicated and inconvenient for non-developers to use directly. An intelligent conversation assistant is needed to help answer various common business questions of users.

[0029] At present, intelligent conversation assistants all adopt the solution of question-and-answer knowledge base, which stores answers to common questions in advance, uses artificial intelligence technology to match corresponding answers to user questions, and then responds to user business questions. However, in actual applications, user business questions often involve complex business query, operation and other functions. It is difficult to support complex business query and operation functions only by processing business problems based on question-and-answer knowledge base. Its processing function is relatively simple, and the business processing effect is biased, which makes it difficult to meet users' complex business processing needs.

[0030] Based on this, a business question and answer processing method is provided in an embodiment of the present application, which can query and operate complex businesses to solve the technical problem that intelligent dialogue assistants are difficult to support and process complex business queries and operations.

[0031] Example:

[0032] Figure 1A flowchart of a business question and answer processing method provided in an embodiment of the present application is given. The business question and answer processing method provided in this embodiment can be executed by a business question and answer processing device, which can be implemented by software and / or hardware. The business question and answer processing device can be composed of two or more physical entities, or it can be composed of one physical entity. Generally speaking, the business question and answer processing device can be a terminal device such as a computer, a mobile phone, a tablet, an intelligent dialogue assistant server, etc.

[0033] The following description is made by taking the business question and answer processing device as an example to perform the business question and answer processing method. Figure 1 , the business question and answer processing method specifically includes:

[0034] S110, receiving a target business question, determining a target question vector value of the target business question, searching a pre-built question-answering database and a function knowledge base based on the target question vector value, wherein the question-answering database pre-configures response data corresponding to different question vectors, and the function knowledge base pre-configures business functions corresponding to different question vectors.

[0035] In the process of business question-answering, this application obtains business questions through business interaction with users, and then answers business questions, and defines the obtained business questions as target business questions. For example, a user asks a specific business question to the intelligent dialogue assistant through a corresponding interface or interface, and first determines the target question vector value of the target business question based on the received target business question.

[0036] The target question vector value can use natural language processing (NLP) technology (such as word embedding, BERT and other pre-trained models) to convert the received natural language question into a vector representation, that is, the target question vector value. This vector value can capture the semantic features of the question and facilitate subsequent matching.

[0037] Optionally, determining a target problem vector value of a target business problem includes:

[0038] The target business problem is input into the specified large model, and the target problem vector value corresponding to the target business problem is calculated based on the specified large model.

[0039] When processing the target business problem, the present application sends the target business problem to the designated large model to obtain the vector value of the problem, and uses the vector value as the basis for searching the corresponding reply data. The designated large model (such as GPT) has the ability to understand and generate natural language, and can capture and understand the semantic information of the text. When the target business problem is input into the designated model, the target business problem can be integrated using a pre-set input template so that the designated large model knows how to process the target business problem. The input template can be "Please convert the following "business problem" into a vector group value". In addition, specific functions can be trained based on the designated large model so that the designated large model can output the corresponding problem vector value based on the target business problem of the input model. The designated large model will encode the input target business problem based on its trained weights and architecture to generate a vector representation, namely the target problem vector value. This vector value can capture the key semantic features of the problem and provide a basis for subsequent problem matching and processing.

[0040] Based on the obtained question vector value, you can search the pre-built question-answer database and function knowledge base to obtain the response data of the target business question.

[0041] Prior to this, refer to Figure 2 When building a question-answering database, the business question is sent to the specified large model to obtain the vector value of the question, and then the vector value, business question and answer data are stored together in the Redis vector database as the basis for question-answering search. In addition, the vector value, business question and answer data can also be stored in the MySQL database as a backup.

[0042] In addition, when building the function knowledge base, when storing a business function, the function knowledge base will add one or more business problem keywords corresponding to the business function, send these business problem keywords to the specified large model to obtain the corresponding vector values, and then store the vector values ​​and function configurations together in the Redis vector database as the basis for function search, and store them in the MySQL database as a backup.

[0043] Based on the above pre-built question-answering database and function knowledge base, the target question vector value can be used to search for response data to provide responses to target business questions.

[0044] S120. Determine first reply data that matches the target question vector value from the question and answer database, and / or determine target function information that matches the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data.

[0045] Reference Figure 4When searching the question-answering database based on the target question vector value, since the question-answering database stores various types of business questions, vector values, and reply data, there is a corresponding mapping relationship between different question vectors and corresponding reply data. Therefore, when searching the question-answering database, the vector closest to the target question vector value (through cosine similarity, Euclidean distance, and other measurement methods) can be found in the question-answering database. The reply data corresponding to the mapping of this vector is the reply data of the target business question, which is defined as the first reply data.

[0046] Similarly, similar to the question-answering database, the function knowledge base stores the mapping relationship between question vectors and business functions. Therefore, when searching the function knowledge base, you can find the vector closest to the target question vector value (through cosine similarity, Euclidean distance and other measurement methods) in the function knowledge base. The business function mapped to this vector is the business function that needs to be called and executed for the target business question, which is defined as the target function information.

[0047] It should be noted that in actual applications, the target business question may require one or both of the question-answering database and the function knowledge base to answer. According to the actual retrieval requirements, when matching information is retrieved from the question-answering database and the function knowledge base, it can be referenced to answer the target business question. Otherwise, if the similarity obtained by the vector value search is low and lower than the set threshold, the corresponding reply data will not be obtained.

[0048] Optionally, executing the specified service based on the target function information and determining the second reply data includes:

[0049] Determine the execution parameters corresponding to the target function information, call the corresponding function interface based on the target function information and the execution parameters to execute the specified business, generate the execution result, and determine the second reply data based on the execution result.

[0050] It should be noted that, since the function knowledge base only provides the target function information required to process the target business problem, it is necessary to execute the target function to determine the second response data according to the execution result.

[0051] Specifically, according to the objective function information, the execution parameters required for specifying the objective function information are first determined. These parameters include business-related data, user input values, system status, etc. The execution parameters can be configured when configuring the objective function information, or they can be configured by subsequent system operation and maintenance personnel. Based on the objective function information and the determined execution parameters, the corresponding function interface is found and called. The specified business logic is executed through the function interface. Such as data query, data processing, data conversion and other operations. After the function is executed, the execution results are collected and generated. The execution results may include data sets, status information, error codes, etc. The execution results are formatted or converted as necessary for subsequent processing and display. Based on the execution results, key information is extracted from them, and organized according to user needs and system specifications to obtain second reply data, which may include business results, status information, user prompts, etc.

[0052] Optionally, determining the execution parameters corresponding to the objective function information includes:

[0053] The objective function information is input into a designated large model, an execution parameter type of the objective function information is output based on the designated large model, and an execution parameter corresponding to the objective function information is determined based on the execution parameter type.

[0054] When determining the execution parameters corresponding to the target function information, a designated large model (such as GPT) can also be used to assist this process. By inputting the target function information into the designated large model. The designated large model will output the execution parameter type required for the target function information based on the input target function information and using its trained semantic understanding and generation capabilities. Taking a business query function that queries the ban history of a specific user as an example, the execution parameter type may include user ID, time range, etc. Similarly, when inputting the target function information into the designated large model, the target function information can be integrated using a pre-set input template so that the designated large model knows how to process the target function information and output the corresponding execution parameter type. The input template can be "Please determine the execution parameter type of the following function "target function information"". In addition, specific functions can be trained on the basis of the designated large model so that the designated large model can output the corresponding execution parameter type based on the target function information of the input model. By using the designated large model to determine the execution parameter type of the target function information and determining the specific execution parameters based on these types, user requests can be processed more intelligently and efficiently, and the performance and user experience of the business question and answer processing system can be improved.

[0055] Optionally, determining the execution parameter corresponding to the target function information based on the execution parameter type includes:

[0056] Generate dialogue information based on the execution parameter type, send the dialogue information to the user end that submitted the target business problem, and receive the execution parameters of the target function information returned by the user end based on the dialogue information.

[0057] Based on the above determined execution parameter types, the intelligent dialogue assistant will generate corresponding dialogue information. These dialogue information may include requests prompting users to enter specific types of parameters, parameter formats and examples, etc. In order to provide a better user experience, natural language generation technology can be used to generate more natural and fluent dialogue content.

[0058] The intelligent dialogue assistant sends the generated dialogue information to the user end that submitted the target business problem. It is displayed through a message box, pop-up window or chat window of the user interface (such as a web page, APP, etc.). The user end will display the dialogue information and prompt the user to enter the required execution parameters. After the user enters the parameters according to the prompt, the intelligent dialogue assistant will receive these inputs. The intelligent dialogue assistant further processes the parameters entered by the user, including parsing, verification and formatting. The processed user input will be determined as the execution parameters of the target function information. Then, the determined execution parameters are used to call the target function, execute the specified business logic, and generate the corresponding execution results.

[0059] For example, refer to Figure 4 , specifying the large model to determine whether more function execution parameters are needed based on the function configuration and the user's conversation content. If more execution parameters are needed, the intelligent conversation assistant will have further conversations with the user and ask the user to enter the execution parameters. After obtaining the function configuration and function parameters to be executed, the intelligent assistant will send the function execution information to the function caller, and the caller will execute the function and return the execution result to the intelligent assistant, thereby constructing the second reply data.

[0060] Optionally, the target function information includes at least one of a business query function, a business update function, or a business execution function. The target function information can be subdivided into multiple types to meet the needs of different business scenarios. Among them, the business query function is mainly used to retrieve information from the database to meet the user's query needs. For example, the user wants to query the ban history, reward record, or account balance of a specific user. In this case, by calling the corresponding business query function and executing the database query according to the corresponding execution parameters (such as user ID, time range, etc.), the query result is finally returned as the reply data. The business update function is used to modify the information in the database in response to the user's request. For example, the user wants to update his personal information, change the password, or adjust the account settings. In this case, the system will call the corresponding business update function and perform the database update operation according to the parameters provided by the user. The update operation may include inserting new records, updating existing records, or deleting records that are no longer needed. In addition to query and update operations, the business execution function also covers other business logic processing. For example, a user requests to perform a specific business task, such as participating in an event lottery. In this case, the system will call the corresponding business execution function and perform a series of operations based on the parameters and business rules provided by the user, and finally generate the execution results and return them to the user.

[0061] S130. Respond to the business question based on the first response data and / or the second response data.

[0062] Finally, based on the first reply data and / or the second reply data determined above, the intelligent dialogue assistant can conduct a dialogue with the user terminal and output the corresponding reply result on the user interaction interface. The reply result can directly display the first reply data and / or the second reply data, or it can be a reply text generated based on the first reply data and / or the second reply data using a fixed template.

[0063] By adding function functions, the intelligent conversation assistant can support various query, update or execution functions. For example, query the award record of a certain time period, update the country code of a certain user, etc. In addition, the function keyword matching mechanism is adopted to only return the functions with the highest matching degree, which can optimize the matching accuracy, improve the execution accuracy, and reduce the computing consumption of the specified large model.

[0064] In addition, the function caller can configure the required query function, update function, execution function and other functional functions according to business needs, and support hot deployment of functions. Adding and updating functions will not affect the operation of existing intelligent assistants. You only need to add, update, and delete functions in the configuration center of the intelligent dialogue assistant platform to use them. It will not affect the deployed intelligent dialogue assistant business and can expand the business functions of the intelligent dialogue assistant.

[0065] Respond to business questions based on the first response data and / or the second response data, including:

[0066] The first reply data and / or the second reply data are input into the designated large model, reply information corresponding to the target business question is output based on the designated large model, and the business question is replied based on the reply information.

[0067] Optionally, when processing business problems, the present application uses a designated large model to analyze and integrate the first response data and / or the second response data to generate response information for the target business problem. Among them, the input data is understood and analyzed by the designated large model. In order to improve the accuracy and relevance of the response to the target business question, the data can be optimized, such as extracting key information, reorganizing data, adding context, etc. In addition, the response data can be reorganized, additional explanations or suggestions can be added, and the format can be adjusted. According to actual needs, the designated large model can adaptively optimize and generate response information to achieve a smoother and more natural business question and answer effect. The response information produced by the designated large model is output to the intelligent dialogue assistant, and can be returned to the user end through the intelligent dialogue assistant to conduct business questions and answers for the target business problem.

[0068] In the above, by receiving the target business question, determining the target question vector value of the target business question, searching the pre-built question and answer database and function knowledge base based on the target question vector value, the question and answer database pre-configures the reply data corresponding to different question vectors, and the function knowledge base pre-configures the business functions corresponding to different question vectors; determining the first reply data that matches the target question vector value from the question and answer database, and / or determining the target function information that matches the target question vector value from the function knowledge base, executing the specified business based on the target function information, and determining the second reply data; replying to the business question based on the first reply data and / or the second reply data. By adopting the above technical means, by combining the corresponding business function to execute the specified business, determining the reply data of the business question to reply to the business question, the business processing function of the intelligent dialogue assistant can be expanded, and the query and operation of complex business functions can be realized through the corresponding business functions, so as to improve the business processing effect and the user's business question and answer experience.

[0069] Based on the above embodiments, Figure 5 A schematic diagram of the structure of a business question and answer processing system provided by this application. Figure 5 The business question and answer processing system provided in this embodiment specifically includes: a search module 21, a matching module 22 and a reply module 23.

[0070] The search module 21 is configured to receive a target business question, determine a target question vector value of the target business question, and search a pre-built question-answer database and a function knowledge base based on the target question vector value, wherein the question-answer database pre-configures reply data corresponding to different question vectors, and the function knowledge base pre-configures business functions corresponding to different question vectors;

[0071] The matching module 22 is configured to determine first reply data matching the target question vector value from the question-answer database, and / or determine target function information matching the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data;

[0072] The reply module 23 is configured to reply to the business question based on the first reply data and / or the second reply data.

[0073] Specifically, executing the specified business based on the target function information and determining the second reply data includes:

[0074] Determine the execution parameters corresponding to the target function information, call the corresponding function interface based on the target function information and the execution parameters to execute the specified business, generate the execution result, and determine the second reply data based on the execution result.

[0075] Wherein, determining the execution parameters corresponding to the objective function information includes:

[0076] The objective function information is input into the specified large model, the execution parameter type of the objective function information is output based on the specified large model, and the execution parameter corresponding to the objective function information is determined based on the execution parameter type.

[0077] Determine the execution parameters corresponding to the target function information based on the execution parameter type, including:

[0078] Generate dialogue information based on the execution parameter type, send the dialogue information to the user end that submitted the target business problem, and receive the execution parameters of the target function information returned by the user end based on the dialogue information.

[0079] Specifically, responding to the business question based on the first response data and / or the second response data includes:

[0080] The first reply data and / or the second reply data are input into the designated large model, reply information corresponding to the target business question is output based on the designated large model, and the business question is replied based on the reply information.

[0081] Specifically, the target function information includes at least one of a service query function, a service update function or a service execution function.

[0082] Specifically, determine the target problem vector value of the target business problem, including:

[0083] The target business problem is input into the specified large model, and the target problem vector value corresponding to the target business problem is calculated based on the specified large model.

[0084] In the above, by receiving the target business question, determining the target question vector value of the target business question, searching the pre-built question and answer database and function knowledge base based on the target question vector value, the question and answer database pre-configures the reply data corresponding to different question vectors, and the function knowledge base pre-configures the business functions corresponding to different question vectors; determining the first reply data that matches the target question vector value from the question and answer database, and / or determining the target function information that matches the target question vector value from the function knowledge base, executing the specified business based on the target function information, and determining the second reply data; replying to the business question based on the first reply data and / or the second reply data. By adopting the above technical means, by combining the corresponding business function to execute the specified business, determining the reply data of the business question to reply to the business question, the business processing function of the intelligent dialogue assistant can be expanded, and the query and operation of complex business functions can be realized through the corresponding business functions, so as to improve the business processing effect and the user's business question and answer experience.

[0085] The business question and answer processing system provided in the embodiment of the present application can be configured to execute the business question and answer processing method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0086] Based on the above practical example, the present application embodiment also provides a business question and answer processing device, referring to Figure 6 , the business question and answer processing device includes: a processor 31, a memory 32, a communication module 33, an input device 34 and an output device 35. The memory, as a computer-readable storage medium, can be configured to store software programs, computer executable programs and modules, such as program instructions / modules corresponding to the business question and answer processing method described in any embodiment of the present application (for example, the search module, matching module and reply module in the business question and answer processing system). The communication module is configured to perform data transmission. The processor executes various functional applications and data processing of the device by running the software programs, instructions and modules stored in the memory, that is, realizes the above-mentioned business question and answer processing method. The input device can be configured to receive input digital or character information, and generate key signal input related to the user settings and function control of the device. The output device may include a display device such as a display screen. The business question and answer processing device provided above can be configured to execute the business question and answer processing method provided in the above embodiment, and has corresponding functions and beneficial effects.

[0087] On the basis of the above embodiments, the embodiments of the present application further provide a non-volatile computer-readable storage medium, wherein the non-volatile computer-readable storage medium stores computer-executable instructions, wherein the computer-executable instructions are configured to execute a business question-and-answer processing method when executed by a computer processor, and the storage medium may be any of various types of memory devices or storage devices. Of course, the non-volatile computer-readable storage medium provided in the embodiments of the present application, whose computer-executable instructions are not limited to the business question-and-answer processing method described above, may also execute related operations in the business question-and-answer processing method provided in any embodiment of the present application.

[0088] On the basis of the above embodiments, the embodiments of the present application also provide a computer program product. The essence of the technical solution of the present application or the part that contributes to the prior art or all or part of the technical solution can be embodied in the form of a software product. The computer program product is stored in a storage medium, and includes a number of instructions for enabling a computer device, a mobile terminal or a processor therein to execute all or part of the steps of the business question and answer processing method described in each embodiment of the present application.

Claims

1. A business question and answer processing method, characterized in that: include: Receive a target business question, determine a target question vector value of the target business question, and search a pre-built question-answer database and a function knowledge base based on the target question vector value, wherein the question-answer database pre-configures reply data corresponding to different question vectors, and the function knowledge base pre-configures business functions corresponding to different question vectors; Determine first reply data matching the target question vector value from the question-answer database, and / or determine target function information matching the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data; The business question is answered based on the first response data and / or the second response data.

2. The business question and answer processing method according to claim 1, characterized in that: The performing of the designated business based on the objective function information to determine the second reply data includes: Determine the execution parameters corresponding to the target function information, call the corresponding function interface based on the target function information and the execution parameters to execute the specified business, generate an execution result, and determine the second reply data based on the execution result.

3. The business question and answer processing method according to claim 2, characterized in that: The determining of the execution parameters corresponding to the objective function information includes: The objective function information is input into a designated large model, an execution parameter type of the objective function information is output based on the designated large model, and an execution parameter corresponding to the objective function information is determined based on the execution parameter type.

4. The business question and answer processing method according to claim 3, characterized in that: The determining the execution parameter corresponding to the objective function information based on the execution parameter type includes: Generate dialogue information based on the execution parameter type, send the dialogue information to the user end that submitted the target business problem, and receive the execution parameters of the target function information returned by the user end based on the dialogue information.

5. The business question and answer processing method according to any one of claims 1 to 4, characterized in that: Responding to the business question based on the first response data and / or the second response data includes: The first reply data and / or the second reply data are input into a designated large model, reply information corresponding to the target business question is output based on the designated large model, and the business question is replied based on the reply information.

6. The business question and answer processing method according to any one of claims 1 to 4, characterized in that: The target function information includes at least one of a service query function, a service update function or a service execution function.

7. The business question and answer processing method according to any one of claims 1 to 4, characterized in that: The determining of the target problem vector value of the target business problem includes: The target business problem is input into a specified large model, and a target problem vector value corresponding to the target business problem is calculated based on the specified large model.

8. A business question and answer processing system, characterized in that: include: A search module is configured to receive a target business question, determine a target question vector value of the target business question, and search a pre-built question-answer database and a function knowledge base based on the target question vector value, wherein the question-answer database is pre-configured with response data corresponding to different question vectors, and the function knowledge base is pre-configured with business functions corresponding to different question vectors; a matching module, configured to determine first reply data matching the target question vector value from the question-answer database, and / or determine target function information matching the target question vector value from the function knowledge base, execute a specified business based on the target function information, and determine second reply data; A reply module is configured to reply to the business question based on the first reply data and / or the second reply data.

9. A business question and answer processing device, characterized in that: include: memory and one or more processors; The memory is configured to store one or more programs; When the one or more programs are executed by the one or more processors, the one or more processors implement the business question and answer processing method as described in any one of claims 1-7.

10. A non-volatile computer-readable storage medium, characterized in that: The non-volatile computer-readable storage medium stores computer-executable instructions, and when the computer-executable instructions are executed by a computer processor, they are configured to execute the business question-and-answer processing method according to any one of claims 1 to 7.

11. A computer program product, characterized in that The computer program product includes instructions, and when the instructions are executed on a computer or a processor, the computer or the processor executes the business question and answer processing method as described in any one of claims 1 to 7.

Images