Business request processing method and device, electronic equipment and storage medium

Abstract

This application relates to the field of data processing technology, applied to fintech and smart healthcare scenarios, and provides a business request processing method, apparatus, electronic device, and storage medium. The method includes: acquiring a business request; extracting structured features from the business request to obtain business structured features; performing semantic vectorization processing on the business structured features to obtain business semantic vectors; matching the business semantic vectors with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain matching results, wherein the intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent; determining at least one target intelligent agent from the intelligent agent cluster based on the matching results; and processing the business request based on the target intelligent agent to obtain a business request processing result. Through the above technical solution, the accuracy of the system output results can be improved, meeting the needs of complex decision-making scenarios.

Description

Technical Field

[0001] The embodiments of this application relate to, but are not limited to, the field of data processing technology, and are applied to financial technology and smart healthcare scenarios, particularly to a business request processing method, apparatus, electronic device, and storage medium. Background Technology

[0002] With the deep integration of large language model technology and BI (Business Intelligence) systems, conversational business intelligence has become the mainstream interaction paradigm. Its typical interaction process is as follows: a centralized large language model parses the user's natural language input, translates it into executable SQL statements, queries the database, and then uses preset templates to render charts or generate simple text summaries. For example, in the fintech field, intelligent interactive systems can be used to query indicator data for financial operations to quickly return detailed financial operation data; in the smart healthcare field, intelligent interactive systems can be used to query data for medical and pharmaceutical operations to quickly return relevant detailed data.

[0003] However, current intelligent interaction systems typically simplify user input into a single data extraction request, lacking an effective distinction between data retrieval intent and analytical intent. When user input includes higher-level business demands such as attribution, diagnosis, or trend prediction, existing systems can only mechanically execute query aggregation, failing to map natural language to corresponding professional business analysis models. This results in existing systems only returning data details without providing in-depth logical analysis, creating a significant semantic gap between user intent and system output. Consequently, the accuracy of the output results is greatly reduced, failing to meet the needs of complex decision-making scenarios. Summary of the Invention

[0004] The following is an overview of the subject matter described in detail herein. This overview is not intended to limit the scope of the claims.

[0005] To address the problems mentioned in the background section, this application provides a business request processing method, apparatus, electronic device, and storage medium. It employs a dynamic mapping mechanism based on business requests and intelligent agent business capabilities to achieve accurate mapping from user intent to complex business analysis logic. This allows for the dynamic invocation of intelligent agents with corresponding professional business capabilities to process business requests, thereby improving the accuracy of system output results and meeting the needs of complex decision-making scenarios.

[0006] In a first aspect, embodiments of this application provide a business request processing method, the business request processing method comprising: Obtain the business request; The business request is subjected to structured feature extraction to obtain the business structured features; The business structured features are semantically vectorized to obtain business semantic vectors; The business semantic vector is matched with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent. At least one target agent is determined from the agent cluster based on the matching results; The business request is processed based on the target intelligent agent to obtain the business request processing result.

[0007] Secondly, embodiments of this application also provide a service request processing apparatus, the service request processing apparatus comprising: The first unit is used to obtain business requests; The second unit is used to extract structured features from the business request to obtain business structured features; The third unit is used to perform semantic vectorization processing on the business structured features to obtain business semantic vectors; The fourth unit is used to match the business semantic vector with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent. The fifth unit is used to determine at least one target agent from the agent cluster based on the matching result; The sixth unit is used to process the business request based on the target intelligent agent and obtain the business request processing result.

[0008] Thirdly, embodiments of this application also provide an electronic device, including: a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the service request processing method described in the first aspect above.

[0009] Fourthly, embodiments of this application also provide a computer-readable storage medium storing computer-executable instructions for performing the business request processing method described in the first aspect above.

[0010] The business request processing method according to the embodiments provided in this application has at least the following beneficial effects: In the process of processing a business request, the business request is first obtained; then, the business request is subjected to structured feature extraction to obtain structured business features; then, the structured business features are subjected to semantic vectorization processing to obtain a business semantic vector; then, the business semantic vector is matched with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result, wherein the intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent; then, at least one target intelligent agent is determined from the intelligent agent cluster based on the matching result; finally, the business request is processed based on the target intelligent agent to obtain the business request processing result. Through the above technical solution, a dynamic mapping mechanism based on business requests and intelligent agent business capabilities is adopted to achieve accurate mapping from user intent to complex business analysis logic, enabling the dynamic invocation of intelligent agents with corresponding professional business capabilities to process business requests, thereby improving the accuracy of the system output results and meeting the needs of complex decision-making scenarios. Attached Figure Description

[0011] The accompanying drawings are used to provide a further understanding of the technical solutions of this application and constitute a part of the specification. They are used together with the embodiments of this application to explain the technical solutions of this application and do not constitute a limitation on the technical solutions of this application.

[0012] Figure 1 This is a schematic diagram of an application environment for a business request processing method according to one embodiment of this application; Figure 2 This is a flowchart illustrating a business request processing method provided in one embodiment of this application; Figure 3 yes Figure 2 A schematic diagram of a specific implementation method of step S200; Figure 4 yes Figure 2 A schematic diagram of a specific implementation of step S300; Figure 5 yes Figure 2 A schematic diagram of a specific implementation of step S400; Figure 6 yes Figure 2 A schematic diagram of a specific implementation of step S500; Figure 7 yes Figure 2 A schematic diagram of a specific implementation of step S600; Figure 8 This is a schematic diagram of a service request processing apparatus provided in one embodiment of this application; Figure 9This is a schematic diagram of an electronic device provided in one embodiment of this application. Detailed Implementation

[0013] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application.

[0014] It should be noted that although functional modules are divided in the device schematic diagram and a logical order is shown in the flowchart, in some cases, the steps shown or described may be performed in a different order than the module division in the device or the order in the flowchart. The terms "first," "second," etc., in the specification, claims, and the aforementioned drawings are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be noted that, unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs. The terminology used herein is for the purpose of describing embodiments of this application only and is not intended to limit this application.

[0015] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0016] AI is a new technical science that studies and develops theories, methods, technologies, and application systems for simulating, extending, and expanding human intelligence. Artificial intelligence is a branch of computer science that attempts to understand the essence of intelligence and produce new intelligent machines that can react in a way similar to human intelligence. Research in this field includes robotics, speech recognition, image recognition, natural language processing, and expert systems. Artificial intelligence can simulate the information processes of human consciousness and thought. Furthermore, artificial intelligence utilizes digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceiving the environment, acquiring knowledge, and using that knowledge to achieve optimal results—the theories, methods, technologies, and application systems available for use.

[0017] Foundational technologies for artificial intelligence generally include sensors, dedicated AI chips, cloud computing, distributed storage, big data processing, operating / interactive systems, and mechatronics. AI software technologies mainly encompass computer vision, robotics, biometrics, speech processing, natural language processing, and machine learning / deep learning.

[0018] Artificial intelligence, or AI, is the theory, method, technology, and application system that uses digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0019] The servers involved in artificial intelligence technology can be standalone servers or cloud servers that provide basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDN), and big data and artificial intelligence platforms.

[0020] This application provides a business request processing method, apparatus, electronic device, and storage medium. In processing a business request, the following steps are taken: First, the business request is acquired; then, structured features are extracted from the business request to obtain structured business features; next, semantic vectorization is performed on the structured business features to obtain a semantic business vector; then, the semantic business vector is matched with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector represents the business processing capability of the corresponding intelligent agent; then, at least one target intelligent agent is determined from the intelligent agent cluster based on the matching result; finally, the business request is processed based on the target intelligent agent to obtain the business request processing result. Through the above technical solution, a dynamic mapping mechanism based on business requests and intelligent agent business capabilities is adopted to achieve accurate mapping from user intent to complex business analysis logic. This allows for the dynamic invocation of intelligent agents with corresponding professional business capabilities to process business requests, thereby improving the accuracy of the system output results and meeting the needs of complex decision-making scenarios.

[0021] The business request processing method provided in this application relates to the field of data processing technology. This method can be applied to a terminal, a server, or software running on either a terminal or a server. In some embodiments, the terminal can be a smartphone, tablet, laptop, desktop computer, etc.; the server can be configured as an independent physical server, a server cluster or distributed system composed of multiple physical servers, or a cloud server providing basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, CDN, and big data and artificial intelligence platforms.

[0022] This application can be used in a wide variety of general-purpose or special-purpose computer system environments or configurations. Examples include: personal computers, server computers, handheld or portable devices, tablet devices, multiprocessor systems, microprocessor-based systems, set-top boxes, programmable consumer electronics, network PCs, minicomputers, mainframe computers, and distributed computing environments including any of the above systems or devices. This application can be described in the general context of computer-executable instructions executed by a computer, such as program modules. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform specific tasks or implement specific abstract data types. This application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In distributed computing environments, program modules can reside in local and remote computer storage media, including storage devices.

[0023] It should be noted that in all specific embodiments of this application, when processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, user permission or consent is obtained first. Furthermore, the collection, use, and processing of this data comply with relevant laws, regulations, and standards. In addition, when embodiments of this application require access to sensitive personal information of users, separate permission or consent from the user is obtained through pop-ups or redirection to confirmation pages. Only after obtaining the user's separate permission or consent is the necessary user-related data required for the proper functioning of these embodiments acquired.

[0024] The business request processing method provided in this application embodiment can be applied to, for example, Figure 1In this application environment, the client communicates with the server via a network. The server can obtain business requests from the client; then, it extracts structured features from the business requests to obtain structured business features; next, it performs semantic vectorization on the structured business features to obtain business semantic vectors; then, it matches the business semantic vectors with multiple capability description vectors corresponding to a pre-defined intelligent agent cluster to obtain matching results. The intelligent agent cluster includes multiple intelligent agents handling different business types, and each capability description vector represents the corresponding intelligent agent's business processing capabilities. Next, based on the matching results, at least one target intelligent agent is determined from the intelligent agent cluster; finally, the business request is processed based on the target intelligent agent to obtain the business request processing result. Through the above technical solution, a dynamic mapping mechanism based on business requests and intelligent agent business capabilities is adopted to achieve accurate mapping from user intent to complex business analysis logic. This allows for the dynamic invocation of intelligent agents with corresponding professional business capabilities to process business requests, thereby improving the accuracy of the system output results and meeting the needs of complex decision-making scenarios. The client can be, but is not limited to, various personal computers, laptops, smartphones, tablets, and portable wearable devices. The server can be implemented using a standalone server or a server cluster consisting of multiple servers. The embodiments of this application will be described in detail below through specific examples.

[0025] like Figure 2 As shown, Figure 2 This is a flowchart illustrating a business request processing method provided in one embodiment of this application. The business request processing method includes the following steps: Step S100: Obtain the business request.

[0026] The business request processing method provided in this application first obtains the business request input by the user during the process of processing the business request. The input methods include, but are not limited to, text input and voice input.

[0027] In fintech scenarios, users input business requests into financial business systems. For example, when bank finance personnel use a financial intelligent interaction system to inquire about the reasons for the decline in financial operating indicators, the bank staff can input the business request into the financial intelligent interaction system in the form of "analyze the reasons for the decline in financial operating indicators" using natural language. The business request then enters the financial intelligent interaction system in the form of business request data.

[0028] In the context of smart healthcare, users input business requests into the medical business system. For example, when bank finance personnel use the medical smart interaction system to inquire about the reasons for the continuous decline in the inventory of drug A, medical personnel can input the business request into the medical smart interaction system in the form of natural language, such as "analyze the reasons for the continuous decline in the inventory of drug A". The business request then enters the medical smart interaction system in the form of business request data.

[0029] It is understandable that, in the process of obtaining business request data, when it involves processing data related to user identity or characteristics, such as user information, user behavior data, user historical data, and user location information, the user's permission or consent will be obtained first. Furthermore, the collection, use, and processing of this data will comply with relevant laws, regulations, and standards. In addition, when this application embodiment needs to obtain sensitive personal information of users, separate permission or consent from the user will be obtained through pop-ups or redirects to confirmation pages. Only after explicitly obtaining the user's separate permission or consent will the necessary user-related data for the normal operation of this application embodiment be obtained.

[0030] Step S200: Extract structured features from the business request to obtain the business structured features.

[0031] The business request processing method provided in this application requires extracting structured features from the business request during the processing process to obtain corresponding business structured features. These features include business information such as the business type, purpose, and object of the request, reflecting the user's intent in making the request. For example, in a financial business scenario, a bank employee can input this business request into a financial intelligent interaction system using natural language, such as "Analyze the reasons for the decline in financial operating indicators." By extracting structured features from this request, the system obtains business structured features, which include information that the business type is data query, the business purpose is attribution analysis, and the business object is the financial department's operating indicators. In a medical business scenario, medical personnel can input this business request into a medical intelligent interaction system using natural language, such as "Analyze the reasons for the continuous decline in the inventory of drug A." By extracting structured features from this request, the system obtains business structured features, which include information that the business type is data query, the business purpose is attribution analysis, and the business object is the inventory of drug A in the pharmacy. The business request processing method provided in this application can achieve user intent recognition through the business structured features extracted from the business request data.

[0032] like Figure 3 As shown, extracting structured features from business requests to obtain structured business features can include the following steps: Step S210: Perform action recognition on the business request to obtain action recognition information; Step S220: Extract business type features from the action recognition information to obtain business operation type features; Step S230: Perform destination identification on the business request to obtain destination identification information; Step S240: Match the destination identification information with the preset standard destination label to obtain the business destination label features; Step S250: Perform entity identification on the business request to obtain entity identification information; Step S260: Extract business object features from entity recognition information to obtain business entity parameter features.

[0033] It is understood that the business request processing method provided in this application embodiment will use a hierarchical identification mechanism to extract structured features of the business request during the process of processing the business request, and obtain business structured features, which include business operation type features, business purpose label features and business entity parameter features.

[0034] It is understandable that, at the action recognition level, action recognition is performed on business requests to obtain action recognition information, and then business type features are extracted from the action recognition information. Based on this, the business operation type features of user requests can be extracted according to the business requests. The business operation type features are used to characterize the user's business operation type, including but not limited to data query, data comparison, data export, data analysis, etc.

[0035] Understandably, at the purpose identification level, business requests are identified to obtain purpose identification information. This information is then matched with pre-defined standard purpose tags to obtain business purpose tag features. For example, the system has a pre-defined standardized business analysis purpose tag system to characterize specific business analysis logic. This tag system includes several different standard purpose tags, such as: T-1 daily monitoring, T-2 DuPont attribution, T-3 cohort analysis, T-4 anomaly diagnosis, T-5 ranking comparison, and T-6 trend prediction. By matching the purpose identification information with the aforementioned standard purpose tags, the most relevant business purpose tag features can be obtained.

[0036] Understandably, at the entity recognition level, entity recognition is performed on business requests to obtain entity recognition information. Then, business object features are extracted from the entity recognition information to obtain business entity parameter features. Based on this, the system can extract business entity parameter features from user requests. These business entity parameter features are used to characterize business objects, including but not limited to organization name, time range, and indicator fields.

[0037] For example, in a fintech scenario, a finance professional inputs a business request into the financial intelligent interaction system: "Diagnose the recent performance of the Second District Department." At the action recognition level, the system can extract the business operation type feature "[Action: Data Analysis]" from the business request. At the purpose recognition level, the system can extract the business purpose label feature "[Purpose: T-4 Anomaly Diagnosis]" from the business request. At the entity recognition level, the system can extract the business entity parameter feature "[Entity:{Institution: Second District Department, Indicator: Performance}]" from the business request.

[0038] For example, in a smart healthcare scenario, medical staff input a business request into the medical intelligent interaction system: "Predict the inventory changes of the pharmacy department in the next three months." At the action recognition level, the system can extract the business operation type feature "[Action: Data Analysis]" from the business request. At the purpose recognition level, the system can extract the business purpose label feature "[Purpose: T-6 Trend Prediction]" from the business request. At the entity recognition level, the system can extract the business entity parameter feature "[Entity:{Institution: Pharmacy Department, Indicator: Inventory}]" from the business request.

[0039] Based on this, by performing hierarchical recognition of user-inputted business requests, the system can transform the fuzzy natural language input by users into multi-dimensional business structured features from three different levels: action layer, purpose layer, and entity layer. This facilitates accurate identification of user intent based on business operation type features, business purpose label features, and business entity parameter features, thereby effectively solving the problem of semantic understanding gaps in existing technologies.

[0040] Step S300: Perform semantic vectorization processing on the business structured features to obtain the business semantic vector.

[0041] It is understood that the business request processing method provided in this application requires semantic vectorization of the business structured features in order to subsequently match the intelligent agent with the corresponding professional business capabilities to process the business request. This results in a business semantic vector, which converts the natural language context input by the user into a high-dimensional semantic feature vector. Subsequently, the system performs relevance matching based on the business semantic vector and the capability description vectors of each atomic intelligent agent in the intelligent agent cluster. This allows the system to match the relevant intelligent agent to process the business request input by the user, which helps to improve the accuracy of the system output results.

[0042] like Figure 4 As shown, semantic vectorization of business structured features to obtain business semantic vectors can include the following steps: Step S310: Perform feature concatenation on the business operation type feature, business purpose label feature, and business entity parameter feature to obtain standard structured instructions; Step S320: Convert the standard structured instructions into a numerical vector to obtain the business semantic vector.

[0043] It is understood that the business request processing method provided in this application embodiment, in the process of processing business requests, performs feature concatenation on business operation type features, business purpose tag features and business entity parameter features to obtain standard structured instructions, and then converts the standard structured instructions into numerical vectors to obtain business semantic vectors.

[0044] For example, in a fintech scenario, finance personnel input a business request into a financial intelligent interaction system: "Diagnose the recent performance of the Second District Department." This yields the business operation type feature "[Action: Data Analysis]", the business purpose tag feature "[Purpose: T-4 Anomaly Diagnosis]", and the business entity parameter feature "[Entity:{Institution:Second District Department, Indicator:Performance}]". These structured business features are then concatenated to obtain a standardized instruction structure: [Action: Data Analysis] + [Purpose:T-4 Anomaly Diagnosis] + [Entity:{Institution:Second District Department, Indicator:Performance}]. This standardized instruction structure is then subjected to numerical vector conversion to obtain a computer-computable business semantic vector.

[0045] For example, in a smart healthcare scenario, medical staff input a business request into the medical intelligent interaction system: "Predict the inventory changes of the pharmacy department in the next three months." This yields the business operation type feature "[Action: Data Analysis]", the business purpose tag feature "[Purpose: T-6 Trend Prediction]", and the business entity parameter feature "[Entity:{Institution: Pharmacy Department, Indicator: Inventory}]". These structured business features are then concatenated to obtain a standardized instruction structure: [Action: Data Analysis] + [Purpose: T-6 Trend Prediction] + [Entity:{Institution: Pharmacy Department, Indicator: Inventory}]. This standardized instruction structure is then subjected to numerical vector transformation to obtain a computer-computable business semantic vector.

[0046] Based on this, by performing numerical vector conversion on the standardized instruction structure, a computer-computable business semantic vector is obtained, thereby enabling the accurate mapping of user-input natural language into a business semantic vector.

[0047] Step S400: Match the business semantic vector with multiple capability description vectors corresponding to the preset intelligent agent cluster to obtain the matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent.

[0048] It is understood that the business request processing method provided in this application, in the process of processing business requests, obtains a matching result by matching the business semantic vector with multiple capability description vectors corresponding to a preset intelligent agent cluster, so that the system can match the relevant intelligent agent to process the business request input by the user.

[0049] Understandably, this application employs an atomic agent cluster design, decomposing the system's business analysis capabilities into multiple independently running agents. Agents with different business processing capabilities can be independently iterated and tested by different development teams without modifying the main control routing code. This modular architecture completely solves the maintenance difficulties caused by code coupling in traditional monolithic large-model architectures. When new business analysis capabilities are needed, it is only necessary to register the new atomic agent in the registry center and update the routing configuration, without reconstructing the overall system structure, greatly reducing the marginal cost of long-term system evolution.

[0050] Understandably, based on the type of business analysis, the system can dynamically match business requests to various atomic agents within the agent cluster. Agents of different business analysis types each undertake independent business processing tasks. These agents include, but are not limited to, attribution analysis agents, trend fitting agents, and ranking comparison agents. Specifically, attribution analysis agents are responsible for attributing various indicators and root cause analysis; trend fitting agents are responsible for time-series data prediction and trend analysis; and ranking comparison agents are responsible for ranking indicators and multi-dimensional data comparison.

[0051] Understandably, each agent is configured with optimized prompt word templates and toolchains only within its specific domain of expertise. For example, the attribution analysis agent internally encapsulates attribution algorithms such as DuPont analysis and indicator contribution decomposition, along with supporting tools. After receiving structured business parameters, it automatically performs attribution calculations, root cause ranking, and contribution analysis, outputting standardized attribution conclusions. The trend fitting agent encapsulates time series prediction algorithms (such as ARIMA and Prophet), receiving parameters such as time granularity and indicator data to complete trend modeling, cycle prediction, and anomaly inflection point identification. The ranking and comparison agent's implementation logic includes built-in comparison algorithms such as weighted comparison, dimensional ranking, and difference calculation. By receiving parameters from multiple entities and multiple indicators, it automatically completes ranking, difference calculation, year-on-year and month-on-month comparisons, and outputs comparison conclusions.

[0052] Understandably, the system matches the business semantic vector with multiple capability description vectors corresponding to the preset intelligent agent cluster to obtain the matching degree calculation result. Based on the matching degree calculation result, the optimal target intelligent agent is dynamically selected so that the optimal target intelligent agent can be used to process the corresponding business request. The analysis accuracy of the target intelligent agent when processing business requests is significantly improved, enabling the intelligent interaction system to output high-value decision suggestions that conform to business logic, rather than simply returning raw data.

[0053] It is understood that the business request processing method provided in this application embodiment allows each intelligent agent to handle only business request tasks within its professional scope during the processing of business requests. This avoids attentional distraction caused by the mixing of contextual information from the user's natural language, thereby significantly reducing the probability of the model generating illusions and ensuring the professionalism and accuracy of business analysis conclusions.

[0054] like Figure 5 As shown, matching the business semantic vector with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain the matching result can include the following steps: Step S410: Calculate the similarity between the business semantic vector and the capability description vector corresponding to the intelligent agent cluster to obtain a similarity score. Step S420: If the similarity score is greater than the preset similarity threshold, the corresponding capability description vector is used as the matching result.

[0055] It is understood that the business request processing method provided in this application adopts a dynamic routing and distribution mechanism based on deep semantic features during the processing of business requests. The business semantic vector can reflect the deep semantic features of the business request, thereby accurately understanding the user's intent. By calculating the similarity between the business semantic vector and the capability description vector corresponding to the intelligent agent cluster, and selecting the target intelligent agent corresponding to the optimal capability description vector based on the similarity calculation result, the corresponding computing resources, i.e. the target intelligent agent, can be matched from the intelligent agent cluster to process the corresponding business request.

[0056] It is understood that the business request processing method provided in this application, during the process of processing a business request, performs vector similarity matching calculation between the business semantic vector and the capability description vectors corresponding to each intelligent agent in the intelligent agent cluster, thereby obtaining a similarity score. The similarity calculation includes, but is not limited to, cosine similarity calculation. Based on the obtained multiple similarity scores, if the similarity score is greater than a preset similarity threshold, the corresponding capability description vector is used as the matching result.

[0057] Based on this, the business request processing method provided in this application embodiment can significantly improve the system architecture flexibility, business response capability, computing resource utilization, and data accuracy.

[0058] Step S500: Determine at least one target agent from the agent cluster based on the matching results.

[0059] It is understood that the business request processing method provided in this application, during the process of processing a business request, allows the system to determine the optimal target intelligent agent for handling the user's business request based on the matched capability description vector. It should be noted that, depending on the complexity of the business request, the matched capability description vector can be one or more. In other words, the system can determine one or more target intelligent agents from the intelligent agent cluster to handle the business request.

[0060] Based on this, the system performs relevance matching between the business semantic vector and the capability description vectors of each atomized agent in the agent cluster. Based on the matching results, the system can match the relevant target agents to process the business requests input by the user, which helps to improve the accuracy of the system output results.

[0061] like Figure 6 As shown, determining at least one target agent from a cluster of agents based on the matching results may include the following steps: Step S510: If the matching result is a capability description vector, the agent corresponding to the capability description vector is taken as the target agent. Step S520: If the matching result is multiple capability description vectors, then the multiple agents corresponding to the multiple capability description vectors are taken as the target agents.

[0062] Understandably, depending on the complexity of the business request, the matched capability description vector can be one or more. That is, for a simple single task, the matching result is a single capability description vector, and the business request is handled by the best corresponding agent; for a complex compound task, the matching result is multiple capability description vectors, and the business request is handled by multiple different corresponding agents.

[0063] Step S600: Based on the target intelligent agent, process the business request and obtain the business request processing result.

[0064] It is understood that in the business request processing method provided in this application embodiment, during the process of processing a business request, the system calls the target intelligent agent to process the business request. The target intelligent agent performs professional analysis and processing based on the business structured characteristics corresponding to the business request, and finally outputs the business request processing result.

[0065] The business request processing method provided in this application adopts a dynamic mapping mechanism based on business requests and intelligent agent business capabilities. Through hierarchical identification of user business requests, it accurately identifies user intent and matches the business semantic vector reflecting the user intent with multiple capability description vectors corresponding to the intelligent agent cluster. This process calls the optimal target intelligent agent to handle the business request, achieving precise mapping from user intent to complex business analysis logic. This avoids query failures due to insufficient semantic understanding when facing ambiguous instructions. Users do not need to master complex database structures or use precise query syntax; they only need to input natural language descriptions, and the system can accurately map them to the corresponding specialized intelligent agents to handle different business requests.

[0066] Compared with existing technologies, this application significantly improves the accuracy of business request processing results analysis by using the optimal target agent to process the corresponding business requests when processing business requests input by users using natural voice input. Furthermore, based on the target agent's ability to professionally analyze and process corresponding business types, it can output high-value decision suggestions that conform to business logic, rather than simply returning raw data.

[0067] like Figure 7 As shown, obtaining the business request processing result based on the target intelligent agent's processing of the business request can include the following steps: Step S610: When there are multiple target agents, determine the call chain strategy for the multiple target agents, wherein the call chain strategy is used to characterize the call order of the multiple target agents; Step S620: According to the call chain strategy, call multiple target intelligent agents to process the corresponding business structured features in the business request and obtain the business request processing result.

[0068] For complex, multi-faceted business requests, the system invokes multiple target agents to handle the request and determines a call chain strategy for these agents to plan their invocation order. Then, based on this call chain strategy, the system uses a parameter pass-through mechanism to precisely transmit the corresponding structured business features from the request to each target agent in the call chain, ultimately outputting the business request processing result.

[0069] For example, in a fintech scenario, a finance professional inputs a business request into a financial intelligent interaction system using natural language: "Analyze the recent performance of the Second District Department and predict its performance changes over the next three months." This business request is a composite request. The system extracts the following structured features from the request: "Analyze the recent performance of the Second District Department," including the business operation type feature "[Action: Data Analysis]", the business purpose tag feature "[Purpose: T-4 Anomaly Diagnosis]", and the business entity parameter feature "[Entity:{Institution:Second District Department, Indicator:Performance}]". It also extracts another structured feature from the request "Predict its performance changes over the next three months," which includes the business operation type feature "[Action: Data Analysis]", the business purpose tag feature "[Purpose: T-6 Trend Prediction]", and the business entity parameter feature "[Entity:{Institution:Second District Department, ...]". The system performs semantic vectorization on the above-mentioned structured business features to obtain their respective business semantic vectors. These business semantic vectors are then matched with multiple capability description vectors corresponding to each agent in the agent cluster to obtain matching results. Based on these matching results, at least two target agents are identified from the agent cluster: the attribution analysis agent and the trend fitting agent. The system processes these agents by calling the attribution analysis agent and the trend fitting agent, and the call chain is determined to be: first, the attribution analysis agent is called, then the trend fitting agent is called. Specifically, the attribution analysis agent is first called to analyze the causes and obtain the analysis results, which are then passed to the trend fitting agent for trend prediction, resulting in the final result, which is then output.

[0070] For example, in a smart healthcare scenario, medical staff input a business request into the medical intelligent interaction system using natural language: "Analyze the latest inventory distribution of the pharmacy department and predict its inventory changes over the next three months." This business request is a composite request. The system extracts the business structured features from the business request "Analyze the latest inventory distribution of the pharmacy department," which includes the business operation type feature "[Action: Data Query]", the business purpose tag feature "[Purpose: T-1 Daily Monitoring]", and the business entity parameter feature "[Entity:{Institution: Pharmacy Department, Indicator: Inventory}]". It also extracts another business structured feature from the business request "Predict its inventory changes over the next three months," which includes the business operation type feature "[Action: Data Analysis]", the business purpose tag feature "[Purpose: T-6 Trend Prediction]", and the business entity parameter feature "[Entity:{Institution: Pharmacy Department, Indicator: Inventory}]". These business structured features are then semantically vectorized to obtain their respective business semantic vectors. The system matches the business semantic vector with multiple capability description vectors corresponding to each agent in the agent cluster to obtain matching results. Based on the matching results, at least two target agents are identified from the agent cluster: the attribution analysis agent and the trend fitting agent. The system processes the data by calling the attribution analysis agent and the trend fitting agent, and the call chain is determined to be: first, the attribution analysis agent is called, and then the trend fitting agent is called. Specifically, the attribution analysis agent is first called to analyze the causes and obtain the analysis results, and then the analysis results are passed to the trend fitting agent for trend prediction to obtain the final result and output it.

[0071] Based on this, the business request processing method provided in this application adopts a standardized user intent layer recognition mechanism, an atomic intelligent agent cluster architecture, and a dynamic routing mechanism based on deep semantic features. This achieves a precise mapping from user natural language to complex business analysis logic, enabling the system to dynamically call intelligent agents with corresponding professional business capabilities to process business requests. This improves the accuracy of the system output results and meets the needs of complex decision-making scenarios.

[0072] In addition, such as Figure 8 As shown, one embodiment of this application also provides a service request processing apparatus, which includes: Unit 810 is used to obtain business requests; The second unit 820 is used to extract structured features from business requests to obtain business structured features; The third unit, 830, is used to perform semantic vectorization processing on the business structured features to obtain business semantic vectors. Unit 440 is used to match the business semantic vector with multiple capability description vectors corresponding to the preset intelligent agent cluster to obtain the matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent. Unit 5, 850, is used to determine at least one target agent from the agent cluster based on the matching results; Unit 6, 860, is used to process business requests based on the target intelligent agent and obtain the business request processing results.

[0073] The specific implementation of this service request processing device is basically the same as the specific embodiment of the above-described service request processing method, and will not be repeated here.

[0074] In addition, such as Figure 9 As shown, one embodiment of this application also provides an electronic device, which includes: a memory 920, a processor 910, and a computer program stored on the memory 920 and executable on the processor 910.

[0075] The processor 910 and memory 920 can be connected via a bus or other means.

[0076] The non-transient software program and instructions required to implement the service request processing method of the above embodiments are stored in the memory 920. When executed by the processor 910, the service request processing method of each of the above embodiments is executed.

[0077] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0078] Furthermore, one embodiment of this application provides a computer-readable storage medium storing computer-executable instructions that are executed by a processor 910 or a controller, for example, by a processor 910 in the above-described device embodiment, which enables the processor 910 to perform the service request processing method in the above-described embodiment.

[0079] The above embodiments can be used in combination, and modules with the same name in different embodiments may be the same or different.

[0080] The foregoing has described specific embodiments of this application; other embodiments are within the scope of the appended claims. In some cases, the actions or steps described in the claims may be performed in a different order than those shown in the embodiments and may still achieve the desired result. Furthermore, the processes depicted in the drawings do not necessarily have to follow the specific or sequential order shown to achieve the desired result. In some embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0081] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the embodiments of apparatus, devices, and computer-readable storage media are basically similar to the method embodiments, so the descriptions are relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0082] The apparatus, device, computer-readable storage medium and method provided in the embodiments of this application are corresponding. Therefore, the apparatus, device and non-volatile computer storage medium also have similar beneficial technical effects as the corresponding method. Since the beneficial technical effects of the method have been described in detail above, the beneficial technical effects of the corresponding apparatus, device and computer storage medium will not be described again here.

[0083] In the 1990s, improvements to a technology could be clearly distinguished as either hardware improvements (e.g., improvements to the circuit structure of diodes, transistors, switches, etc.) or software improvements (improvements to the methodology). However, with technological advancements, many methodological improvements today can be considered direct improvements to the hardware circuit structure. Designers almost always obtain the corresponding hardware circuit structure by programming the improved methodology into the hardware circuit. Therefore, it cannot be said that a methodological improvement cannot be implemented using hardware physical modules. For example, a Programmable Logic Device (PLD) (such as a Field Programmable Gate Array (FPGA)) is such an integrated circuit whose logic function is determined by the user programming the device. Designers can program a digital system themselves to "integrate" it onto a PLD, without needing chip manufacturers to design and manufacture dedicated integrated circuit chips. Moreover, nowadays, instead of manually manufacturing integrated circuit chips, this programming is mostly implemented using "logic compiler" software, which is similar to the software compiler used when writing program development code. The original code before compilation must also be written in a specific programming language, which is called a Hardware Description Language (HDL). There is not just one HDL, but many kinds, such as ABEL (Advanced Boolean Expression Language), AHDL (Altera Hardware Description Language), Confluence, CUPL (Cornell University Programming Language), HDCal, JHDL (Java Hardware Description Language), Lava, Lola, MyHDL, PALASM, RHDL (Ruby Hardware Description Language), etc. Currently, the most commonly used are VHDL (Very-High-Speed ​​Integrated Circuit Hardware Description Language) and Verilog. Those skilled in the art should also understand that by simply performing some logic programming on the method flow using the aforementioned hardware description languages ​​and programming it into an integrated circuit, the hardware circuit that implements the logic method flow can be easily obtained.

[0084] The controller can be implemented in any suitable manner. For example, it can take the form of a microprocessor or processor and a computer-readable medium storing computer-readable program code (e.g., software or firmware) executable by the (micro)processor, logic gates, switches, application-specific integrated circuits (ASICs), programmable logic controllers, and embedded microcontrollers. Examples of controllers include, but are not limited to, the following microcontrollers: ARC625D, Atmel AT91SAM, Microchip PIC18F26K20, and Silicon Labs C8051F320. A memory controller can also be implemented as part of the control logic of the memory. Those skilled in the art will also recognize that, in addition to implementing the controller in purely computer-readable program code form, the same functionality can be achieved by logically programming the method steps to make the controller take the form of logic gates, switches, ASICs, programmable logic controllers, and embedded microcontrollers. Therefore, such a controller can be considered a hardware component, and the means included therein for implementing various functions can also be considered as structures within the hardware component. Alternatively, the means for implementing various functions can be considered as both software modules implementing the method and structures within the hardware component.

[0085] The systems, devices, modules, or units described in the above embodiments can be implemented by computer chips or entities, or by products with certain functions. A typical implementation device is a computer. Specifically, a computer can be, for example, a personal computer, laptop computer, cellular phone, camera phone, smartphone, personal digital assistant, media player, navigation device, email device, game console, tablet computer, wearable device, or any combination of these devices.

[0086] For ease of description, the above apparatus is described by dividing it into various functional units. Of course, in implementing the embodiments of this application, the functions of each unit can be implemented in one or more software and / or hardware.

[0087] Those skilled in the art will understand that embodiments of this application can be provided as methods, systems, or computer program products. Therefore, embodiments of this application can take the form of entirely hardware embodiments, entirely software embodiments, or embodiments combining software and hardware aspects. Furthermore, embodiments of this application can take the form of computer program products implemented on one or more computer-usable storage media (including but not limited to disk storage, CD-ROM, optical storage, etc.) containing computer-usable program code.

[0088] This specification is described with reference to flowchart illustrations and / or block diagrams of methods, apparatus (systems), and computer program products according to embodiments of this application. It will be understood that each block of the flowchart illustrations and / or block diagrams, and combinations of blocks in the flowchart illustrations and / or block diagrams, can be implemented by computer program instructions. These computer program instructions can be provided to a processor of a general-purpose computer, special-purpose computer, embedded processor, or other programmable data processing apparatus to produce a machine, such that the instructions, which execute via the processor of the computer or other programmable data processing apparatus, generate instructions for implementing the flowchart... Figure 1 One or more processes and / or boxes Figure 1 A device that provides the functions specified in one or more boxes.

[0089] These computer program instructions may also be stored in a computer-readable storage medium that can direct a computer or other programmable data processing device to function in a particular manner, such that the instructions stored in the computer-readable storage medium produce an article of manufacture including instruction means, which are implemented in a process Figure 1 One or more processes and / or boxes Figure 1 The function specified in one or more boxes.

[0090] These computer program instructions may also be loaded onto a computer or other programmable data processing equipment to cause a series of operational steps to be performed on the computer or other programmable equipment to produce a computer-implemented process, thereby providing instructions that execute on the computer or other programmable equipment for implementing the process. Figure 1 One or more processes and / or boxes Figure 1 The steps of the function specified in one or more boxes.

[0091] In a typical configuration, a computing device includes one or more processors (CPU), input / output interfaces, network interfaces, and memory.

[0092] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash memory (FlashRAM). Memory is an example of computer-readable media.

[0093] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information by any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, magnetic disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0094] It should also be noted that the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0095] In this application embodiment, "at least one" refers to one or more, and "more than one" refers to two or more. "And / or" describes the relationship between related objects, indicating that three relationships can exist. For example, A and / or B can represent the existence of A alone, A and B simultaneously, or B alone. A and B can be singular or plural. The character " / " generally indicates that the preceding and following related objects are in an "or" relationship. "At least one of the following" and similar expressions refer to any combination of these items, including any combination of singular or plural items. For example, at least one of a, b, and c can represent: a, b, c, a and b, a and c, b and c, or a and b and c, where a, b, and c can be single or multiple.

[0096] The embodiments of this application can be described in the general context of computer-executable instructions, such as program modules, that are executed by a computer. Generally, program modules include routines, programs, objects, components, data structures, etc., that perform a specific task or implement a specific abstract data type. The embodiments of this application can also be practiced in distributed computing environments where tasks are performed by remote processing devices connected via a communication network. In a distributed computing environment, program modules can reside in local and remote computer storage media, including storage devices.

[0097] The various embodiments in this application are described in a progressive manner. Similar or identical parts between embodiments can be referred to mutually. Each embodiment focuses on describing the differences from other embodiments. In particular, the system embodiments are basically similar to the method embodiments, so the description is relatively simple; relevant parts can be referred to the descriptions of the method embodiments.

[0098] It should be noted that any AI models, software tools, or components not belonging to this company appearing in the embodiments of this application are merely illustrative examples and do not represent actual use. All user personal information involved in the embodiments of this application has been authorized (with the knowledge and consent) by the relevant parties or has been fully authorized by all parties, and the executing entity may obtain it through various legal and compliant means. The collection, storage, use, processing, transmission, provision, and disclosure of the information, data, and signals involved all comply with relevant laws and regulations and do not violate public order and good morals.

[0099] The above description is merely an embodiment of this application and is not intended to limit this application. Various modifications and variations can be made to this application by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principle of this application should be included within the scope of the claims of this application.

Claims

1. A business request processing method, characterized in that, The method includes: Obtain the business request; The business request is subjected to structured feature extraction to obtain the business structured features; The business structured features are semantically vectorized to obtain business semantic vectors; The business semantic vector is matched with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent. At least one target agent is determined from the agent cluster based on the matching results; The business request is processed based on the target intelligent agent to obtain the business request processing result.

2. The business request processing method according to claim 1, characterized in that, The business structured features include business operation type features, business purpose label features, and business entity parameter features. The process of extracting structured features from the business request to obtain the business structured features includes: The service request is subjected to action recognition to obtain action recognition information; The action recognition information is subjected to business type feature extraction to obtain the business operation type feature; The purpose of the service request is identified to obtain purpose identification information; The purpose identification information is matched with a preset standard purpose label to obtain the business purpose label feature; Entity identification is performed on the service request to obtain entity identification information; The entity identification information is subjected to business object feature extraction to obtain the business entity parameter features.

3. The business request processing method according to claim 2, characterized in that, The step of semantic vectorization of the business structured features to obtain business semantic vectors includes: The business operation type feature, the business purpose label feature, and the business entity parameter feature are concatenated to obtain a standard structured instruction. The standard structured instructions are converted into numerical vectors to obtain the business semantic vector.

4. The business request processing method according to claim 1, characterized in that, The step of matching the business semantic vector with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result includes: The similarity score is obtained by calculating the similarity between the business semantic vector and the capability description vector corresponding to the intelligent agent cluster. If the similarity score is greater than a preset similarity threshold, the corresponding capability description vector is used as the matching result.

5. The business request processing method according to claim 4, characterized in that, The step of determining at least one target agent from the agent cluster based on the matching result includes: If the matching result is a capability description vector, the agent corresponding to the capability description vector shall be taken as the target agent. If the matching result is multiple capability description vectors, then the multiple agents corresponding to the multiple capability description vectors are taken as the target agent.

6. The business request processing method according to claim 1, characterized in that, The process of processing the business request based on the target intelligent agent to obtain the business request processing result includes: When there are multiple target agents, a call chain strategy for the multiple target agents is determined, wherein the call chain strategy is used to characterize the call order of the multiple target agents; According to the call chain strategy, multiple target agents are invoked to process the corresponding business structured features in the business request, and the business request processing result is obtained.

7. The business request processing method according to any one of claims 1 to 6, characterized in that, The intelligent agents include attribution analysis intelligent agents, trend fitting intelligent agents, and ranking comparison intelligent agents; The attribution analysis agent is used to process index data attribution mining tasks. The trend fitting agent is used to process time series data trend analysis tasks. The sorting and comparison agent is used to handle the task of sorting and comparing indicator data.

8. A service request processing apparatus, characterized in that, The service request processing device includes: The first unit is used to obtain business requests; The second unit is used to extract structured features from the business request to obtain business structured features; The third unit is used to perform semantic vectorization processing on the business structured features to obtain business semantic vectors; The fourth unit is used to match the business semantic vector with multiple capability description vectors corresponding to a preset intelligent agent cluster to obtain a matching result. The intelligent agent cluster includes multiple intelligent agents that process different business types, and each capability description vector is used to characterize the business processing capability of the corresponding intelligent agent. The fifth unit is used to determine at least one target agent from the agent cluster based on the matching result; The sixth unit is used to process the business request based on the target intelligent agent and obtain the business request processing result.

9. An electronic device, comprising: A memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, when the processor executes the computer program, it implements the service request processing method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing computer-executable instructions, characterized in that, The computer-executable instructions are used to execute the business request processing method according to any one of claims 1 to 7.

Images