Service system access method and device, computer device and storage medium
By using a unified standard communication protocol and natural language command recognition, the system solves the problems of inefficiency and accuracy in accessing enterprise business systems, and realizes an efficient and accurate multimodal business access system, reducing learning costs and service integration complexity.
Patent Information
- Application Number
- CN202511925867.2
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-12-19
- Publication Date
- 2026-05-01
AI Technical Summary
Existing technologies suffer from inefficiency, inaccurate results, and high implementation costs when accessing enterprise business systems. In particular, they are difficult to adapt interfaces between heterogeneous systems, have weak multimodal resource processing capabilities, and have low application value for independent large models.
A unified access framework for different business services is implemented by adopting a unified standard communication protocol (such as MCP, gRPC, RESTful API), target service identification is performed by combining natural language commands, and retrieval accuracy is improved by using multi-strategy fusion retrieval tools.
It has achieved an easy-to-use, efficient, and accurate multimodal business access system, reducing learning costs and service integration complexity, and improving the efficiency and quality of business system access.
Smart Images

Figure CN121957730A_ABST
Abstract
Description
Business system access methods, devices, computer devices and storage media Technical Field
[0001] This disclosure relates to information retrieval technology, and more particularly to a business system access method, apparatus, computer device, and storage medium. Background Technology
[0002] In the field of intelligent access to enterprise business systems, it is usually achieved through technologies such as API calls, single RAG retrieval systems, and independent large model applications. Although each has formed its own technical system, they all have significant limitations and are difficult to meet the needs of enterprises for efficient, intelligent, and integrated business retrieval.
[0003] API calls are a classic way for business systems to interact. Developers need to have a precise grasp of the interface definition, parameter format and calling logic to access the functions. The learning cost is high, the service calling process is cumbersome, and the interface is difficult to adapt to heterogeneous systems, resulting in a serious lack of flexibility.
[0004] Single RAG retrieval systems focus on retrieving document-type resources and are mostly limited to plain text scenarios, with weak processing capabilities for multimodal business resources such as video surveillance, flowcharts, and page renderings.
[0005] Independent large model applications, with their powerful semantic understanding capabilities, achieve mapping and matching between natural language and business requirements. However, the semantic recognition results are often processed independently of the business service system. Since the configurations of different business service systems are not the same, accurate and efficient service scheduling cannot be performed, resulting in low application value.
[0006] In summary, the business system access process suffers from problems such as low efficiency, inaccurate results, and high implementation costs. Summary of the Invention
[0007] To overcome the problems existing in related technologies, this disclosure provides a business system access method, apparatus, computer device, and storage medium. Based on a unified standard communication protocol, a unified access framework for different business services is implemented, and target service identification is completed by combining natural language commands. This broadens the application scenarios and improves the efficiency and quality of business system access, solving the problems of low efficiency, inaccurate results, and high implementation costs in the process of business system access.
[0008] According to a first aspect of the embodiments of this disclosure, a business system access method is provided, comprising: receiving a natural language instruction; filtering a target service matching the natural language instruction from an available service list, the available service list including a plurality of business services and description information corresponding to each business service; and initiating a call request to the target service based on a standard communication protocol.
[0009] Furthermore, the method also includes: performing standardized registration of multiple business services on a registration management server, recording the name and description information of each business service; configuring a unified service discovery interface for multiple business services, and receiving call requests based on the standard communication protocol.
[0010] Furthermore, the step of filtering the target service matching the natural language instruction from the list of available services includes: obtaining the list of available services through the service discovery interface based on the standard communication protocol; parsing the natural language instruction to obtain the user intent; and determining the matching business service from the list of available services as the target service according to the user intent.
[0011] Furthermore, the step of initiating a call request to the target service based on the standard communication protocol includes: initiating the call request to the target service through a unified business execution interface based on the standard communication protocol.
[0012] Furthermore, the method also includes: obtaining a retrieval tool corresponding to the target service, wherein the retrieval tool includes at least any one or more of the following: a video surveillance retrieval tool for processing and retrieving video surveillance resources, a flowchart retrieval tool for processing and retrieving business approval flowcharts, a page effect diagram retrieval tool for processing and retrieving user interface and page design diagrams, and an application interface retrieval tool for processing and retrieving the operation interface of the actual application system; and invoking the retrieval tool to perform multi-strategy fusion retrieval to obtain retrieval results.
[0013] Furthermore, the step of calling the retrieval tool to perform multi-strategy fusion retrieval and obtain retrieval results includes: calling the retrieval tool to obtain at least one original retrieval result; analyzing each of the original retrieval results according to the matching strategy, and taking the original retrieval result with the highest matching degree with the target service as the retrieval result. The matching strategy includes at least any one or more of the following: complete matching strategy, partial matching strategy, semantic matching strategy, fuzzy matching strategy, and numerical matching strategy for the target service.
[0014] Furthermore, the method also includes: obtaining the data type of the retrieval result of the target service request; and encapsulating the retrieval result generated in response to the call request according to the data type and feeding it back based on the standard communication protocol.
[0015] According to a second aspect of the embodiments of this disclosure, a business system access device is provided, comprising: a global instruction receiving module for receiving natural language instructions; a service matching module for filtering a target service that matches the natural language instructions from an available service list, the available service list including multiple business services and description information corresponding to each business service; and a service routing module for initiating a call request to the target service based on a standard communication protocol.
[0016] According to a third aspect of the embodiments of the present disclosure, a computer apparatus is provided, comprising: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute the above-described business system access method.
[0017] According to a fourth aspect of the embodiments of the present disclosure, a non-transitory computer-readable storage medium is provided, wherein when instructions in the storage medium are executed by a processor of a computer device, the computer device is enabled to perform the above-described business system access method.
[0018] The technical solutions provided by the embodiments of this disclosure can include the following beneficial effects: receiving natural language instructions, filtering target services that match the natural language instructions from an available service list, wherein the available service list includes multiple business services and description information corresponding to each business service; and then initiating a call request to the target service based on a standard communication protocol. By recognizing the user's natural language intent, routing business services based on a unified standard communication protocol, and managing different business services through a unified platform, this solves the problems of low efficiency, inaccurate results, and high implementation costs in the process of accessing business systems, and realizes a multimodal business access system that is easy to use, has high response efficiency, and accurate retrieval.
[0019] It should be understood that the above general description and the following detailed description are exemplary and explanatory only, and are not intended to limit this disclosure. Attached Figure Description
[0020] The accompanying drawings, which are incorporated in and form a part of this specification, illustrate embodiments consistent with this disclosure and, together with the description, serve to explain the principles of this disclosure.
[0021] Figure 1 is a flowchart illustrating a business system access method according to an exemplary embodiment.
[0022] Figure 2 is a flowchart illustrating a method for filtering target services according to an exemplary embodiment.
[0023] Figure 3 is a flowchart illustrating another business system access method according to an exemplary embodiment.
[0024] Figure 4 is a flowchart illustrating another business system access method according to an exemplary embodiment.
[0025] Figure 5 is a retrieval flowchart illustrating an exemplary embodiment.
[0026] Figure 6 is a flowchart illustrating another business system access method according to an exemplary embodiment.
[0027] Figure 7 is a flowchart illustrating another business system access method according to an exemplary embodiment.
[0028] Figure 8 is a block diagram illustrating a business system access device according to an exemplary embodiment. Detailed Implementation
[0029] Exemplary embodiments will now be described in detail, examples of which are illustrated in the accompanying drawings. When the following description relates to the drawings, unless otherwise indicated, the same numerals in different drawings denote the same or similar elements. The embodiments described in the following exemplary embodiments do not represent all embodiments consistent with this disclosure. Rather, they are merely examples of apparatuses and methods consistent with some aspects of this disclosure as detailed in the appended claims.
[0030] API calls are a classic way for business systems to interact. Developers need to have a precise grasp of the interface definition, parameter format and calling logic to access the functions. The learning cost is high, the service calling process is cumbersome, and the interface is difficult to adapt to heterogeneous systems, resulting in a serious lack of flexibility.
[0031] Single RAG retrieval systems focus on retrieving document-type resources and are mostly limited to plain text scenarios, with weak processing capabilities for multimodal business resources such as video surveillance, flowcharts, and page renderings.
[0032] Independent large model applications, with their powerful semantic understanding capabilities, achieve mapping and matching between natural language and business requirements. However, the semantic recognition results are often processed independently of the business service system. Since the configurations of different business service systems are not the same, accurate and efficient service scheduling cannot be performed, resulting in low application value.
[0033] To address the aforementioned issues, embodiments of this disclosure provide a business system access method, apparatus, computer device, and storage medium. Based on a unified standard communication protocol, a unified access framework for different business services is implemented. Furthermore, natural language commands are used to identify the target service, broadening the application scenarios and improving the efficiency and quality of business system access. This solves the problems of low efficiency, inaccurate results, and high implementation costs in the business system access process.
[0034] An exemplary embodiment of this disclosure provides a business system access method. The process of using this method to perform semantic routing on user natural language instructions to complete business access is shown in Figure 1, including: step 101, receiving natural language instructions.
[0035] In this step, natural language commands input by the user are received. Natural language commands are unstructured expressions initiated by the user to the business system that conform to everyday language habits. Users do not need to understand cumbersome information such as API interface definitions and parameter formats; they can trigger the business system access process through everyday language expressions.
[0036] According to one exemplary implementation, all business services of a business system share a single access point, accessible to users or clients. Users issue access commands to different business services through this same access point. On the business system side, all access commands from users / clients are received through the same access point and processed uniformly by intelligent decision-making entities such as a large-model semantic routing layer.
[0037] Step 102: Filter the target service that matches the natural language instruction from the list of available services.
[0038] In this step, the various business services supported by the business system are pre-registered on the registration management server, and the names and descriptions of the business services are stored on the registration management server. Upon receiving a natural language instruction, all registered business services are retrieved, and the business service to be invoked is determined based on the user intent recognition result. This step is specifically shown in Figure 2, including: Step 201, obtaining the list of available services through the service discovery interface based on the standard communication protocol.
[0039] According to one exemplary implementation, the standard communication protocol is the Model Context Protocol (MCP). The registration management server is an MCP server, supporting MCP-based communication.
[0040] According to one exemplary implementation, the standard communication protocol may also be a Remote Procedure Call (gRPC) protocol, a RESTful API, etc.
[0041] Upon receiving a natural language command from the user, an asynchronous processing flow is invoked. First, the names and descriptions of all registered business services on the registration management server are compiled into a structured prompt text, i.e., an available service list. This available service list includes multiple business services and their corresponding descriptions.
[0042] Step 202: Parse the natural language instructions to obtain the user's intent.
[0043] In this step, the large model application is invoked, taking the user's natural language commands and a list of available services as input. Timeout limits can also be set to ensure timely response. The large model application uses deep semantic understanding of the natural language commands and the descriptive information of each business service in the list of available services to analyze and obtain the true user intent.
[0044] According to one exemplary implementation, the user intent includes at least one or more of the following: business objective, i.e., the business operation content that the user ultimately intends to achieve; constraints, specific conditions associated with the business objective, such as document number, equipment code, etc., so as to enable accurate routing based on semantics later; and other requirements, such as data format requirements for result feedback.
[0045] According to one exemplary implementation, the large-scale program can be a BERT model, an ERNIE model, or an LLM model.
[0046] Step 203: Based on the user's intent, determine a matching business service from the list of available services as the target service.
[0047] After obtaining the user intent returned by the large model program, the user intent is further parsed to identify and determine the business service that best matches the user intent as the target service.
[0048] Step 103: Initiate a request to call the target service based on the standard communication protocol.
[0049] In this step, based on the standard communication protocol, a call request to the target service is initiated through a unified business execution interface.
[0050] The business execution interface is the interface for calling specific business services. All calls and services are executed and results are returned through this interface. Execution engine entities such as Multimodal Retrieval Enhanced Generation (RAG) receive call requests through this interface to complete information retrieval and result feedback.
[0051] An exemplary embodiment of this disclosure also provides a method in which multiple business services are registered on a unified registration management server. The registration process is shown in Figure 3, including: step 301, performing standardized registration of multiple business services on the registration management server.
[0052] In this step, various different business services are registered and declared on a unified registration management server. This registration management server can be an MCP service registration management server. The name and description information of each business service are stored in this registration management server.
[0053] According to one exemplary implementation, the name may be a unique identifier for the business service. The description information includes at least one or more of the following: functional description information, such as the purpose of the business service, input parameters, etc.; communication configuration information, such as the deployment address of the business service node, communication protocol, and returned result data format, etc.
[0054] According to an exemplary implementation, the following business services are registered on the registration management server: a service for opening a specified video surveillance by name, a tool for viewing business approval flowcharts, a function for viewing page mockups, and an interface for obtaining defect ticket data, etc.
[0055] Step 302: Configure a unified service discovery interface for the multiple business services, and receive call requests based on the standard communication protocol.
[0056] In this step, the registration management server provides a unified service discovery interface for accessing relevant data of registered business services. All business services can be discovered and invoked by the upper layer in a standardized manner.
[0057] Once the business service registration is complete, the natural language commands issued by the user can be parsed to determine the business service that needs to be invoked.
[0058] Step 303: Receive natural language instructions.
[0059] Step 304: Filter the list of available services to find the target service that matches the natural language instruction.
[0060] Step 305: Initiate a request to call the target service based on the standard communication protocol.
[0061] The implementation principles of steps 303 to 305 are the same as those of steps 101 to 103, and will not be repeated here.
[0062] According to one exemplary implementation, new registration services can be added, or existing registration services that have been registered to the registration management service can be deregistered or have their descriptions modified.
[0063] An exemplary embodiment of this disclosure also provides a business system access method. After determining the business service to be invoked according to the natural language instruction and issuing an invocation request, the process of retrieving according to the invocation request is shown in Figure 4, including: step 401, performing standardized registration of multiple business services on the registration management server.
[0064] Step 402: Configure a unified service discovery interface for the multiple business services, and receive call requests based on the standard communication protocol.
[0065] Step 403: Receive natural language instructions.
[0066] Step 404: Filter the target service that matches the natural language instruction from the list of available services.
[0067] Step 405: Initiate a request to call the target service based on the standard communication protocol.
[0068] The implementation principles of steps 401 to 405 are the same as those of steps 301 to 305, and will not be repeated here.
[0069] After issuing the call request, in response to the call request, the execution engine entities such as the multimodal RAG detection layer make specific business service calls to obtain the detection results. The specific process is as follows.
[0070] Step 406: Obtain the search tool corresponding to the target service.
[0071] In this disclosure, different business services correspond to different search tools, and each search tool is designed for a specific type of data and / or business scenario.
[0072] The retrieval tools include at least one or more of the following: a video surveillance retrieval tool for processing and retrieving video surveillance resources; a flowchart retrieval tool for processing and retrieving business approval flowcharts; a page effect diagram retrieval tool for processing and retrieving user interface and page design diagrams; and an application interface retrieval tool for processing and retrieving the operation interface of the actual application system.
[0073] Step 407: Call the search tool to perform multi-strategy fusion search and obtain search results.
[0074] In this embodiment, the retrieval tool supports multimodal retrieval, and the supported data types include at least one or more of the following: text, images, and videos.
[0075] In this step, the retrieval tool performs multimodal retrieval based on the target service and generates retrieval results. To achieve high-precision matching, the retrieval tool adopts a multi-strategy fusion retrieval algorithm to obtain multiple original retrieval results based on multiple different retrieval strategies. The specific process is shown in Figure 5, including: Step 501, calling the retrieval tool to obtain at least one original retrieval result.
[0076] In this step, after calling the search tool to perform the search, at least one original search result is obtained.
[0077] According to one exemplary implementation, this step can yield multiple original search results, and the degree of matching between different original search results and user intent varies.
[0078] Step 502: Analyze each of the original search results according to the matching strategy, and take the original search result with the highest matching degree with the target service as the search result.
[0079] The matching strategy includes at least one or more of the following: a complete matching strategy, a partial matching strategy, a semantic matching strategy, a fuzzy matching strategy, and a digital matching strategy for the target service.
[0080] In this step, when calculating the matching degree / overall similarity between the user's intent and the original search results, the original search results are evaluated using a matching strategy, and the original search results with high evaluation (i.e., high matching degree with the target service) are selected as the final search results. According to an exemplary implementation, the most relevant results are obtained by weighting different search strategy weights. For example, higher weights are assigned to matching high-precision numbers such as document numbers and device codes. Thus, when evaluating multiple original search results, those that meet the constraint of document number or device code will receive a higher weighted value.
[0081] According to one exemplary implementation, algorithms such as cosine similarity and edit distance can be used to enhance the current matching strategy.
[0082] According to one exemplary implementation, the system can cache some intermediate data or the results of frequently requested requests during the retrieval process. Subsequent identical or similar calls can directly utilize the cached results, further shortening the response time. The caching scheme can be implemented using distributed caches such as Redis and Memcached.
[0083] An exemplary embodiment of this disclosure also provides a business system access method. The process of using this method to feed back the retrieval results to the user / client is shown in Figure 6, including: step 601, performing standardized registration of multiple business services on the registration management server.
[0084] Step 602: Configure a unified service discovery interface for the multiple business services, and receive call requests based on the standard communication protocol.
[0085] Step 603: Receive natural language instructions.
[0086] Step 604: Filter the target service that matches the natural language instruction from the list of available services.
[0087] Step 605: Initiate a request to call the target service based on the standard communication protocol.
[0088] Step 606: Obtain the search tool corresponding to the target service.
[0089] Step 607: Call the search tool to perform multi-strategy fusion search and obtain search results.
[0090] The implementation principles of steps 601 to 607 are the same as those of steps 401 to 407, and will not be repeated here.
[0091] After obtaining the search results, the search results can be returned to the user or client that initiated the business request. The specific process is as follows: Step 608: Obtain the data type of the search results of the target service request.
[0092] Step 609: Encapsulate the search results generated in response to the call request according to the data type and feed them back based on the standard communication protocol.
[0093] In this step, based on the parsing results of the natural language instructions, the user's requirements for the search result format, search success / failure status, response time, and other information are determined, and the search results are encapsulated. Then, based on standard communication protocols, the encapsulated data packets are fed back, and client-side terminals can output the received search results, such as extracting and previewing, downloading, or outputting to a display device.
[0094] An exemplary embodiment of this disclosure also provides a business access system, the architecture of which is shown in Figure 7, including: a first layer: an MCP service registration management layer; a second layer: a large model semantic routing layer; and a third layer: a multimodal RAG retrieval layer.
[0095] 1. The first layer is the system's basic service set, which can be implemented through a registration management server (such as an MCP server). Various different business services are centrally registered and declared on a unified MCP server. Each business service is assigned a clear name and functional description, which is stored on the MCP server. Examples include: a service for opening a specific video surveillance camera by name, a tool for viewing business approval flowcharts, a function for viewing page mockups, and an interface for retrieving defect ticket data. This layer enables all business capabilities to be discovered and invoked by the upper layers in a standardized manner.
[0096] 2. The second layer is the system's intelligent decision-making center, which is responsible for understanding user intentions and assigning tasks.
[0097] When the system starts up, it will automatically load a list of all registered and available services and their function descriptions from the MCP server to achieve service discovery.
[0098] When a user inputs a natural language command, the system invokes an asynchronous processing flow. This flow first constructs a structured prompt text from the descriptions of all available business services, then calls the large model application interface and sets a timeout limit to ensure timely response. The large model program analyzes the user's true intent based on a deep semantic understanding of the user input and all service descriptions.
[0099] 3. The third layer is the system's execution engine, responsible for accurate information retrieval and presentation based on the routing results. It includes an architecture comprised of multiple retrieval tools, each targeting a specific type of data and / or business scenario. For example: a video surveillance retrieval tool specifically designed for processing and retrieving video surveillance resources.
[0100] Flowchart search tool: Specifically designed for processing and searching business approval flowcharts.
[0101] Page mockup search tool: Specifically designed for processing and retrieving user interface and page design mockups.
[0102] Application Interface Search Tool: Specifically designed for processing and searching the user interfaces of actual application systems.
[0103] Semantic intent recognition: When a user makes a request in natural language, the client immediately activates the semantic routing layer upon receiving the input, calling a large model to perform in-depth analysis of the user's statement in order to identify their intent and match the most suitable service.
[0104] Intelligent service routing: Based on the conclusions drawn from semantic analysis, the system routes task requests and invokes the specific business service matched in the first step through a unified MCP communication protocol.
[0105] When the MCP server starts up, it registers the various business services it hosts with the registry center. Clients automatically discover and establish connections with these available business services by reading a unified configuration file.
[0106] The specific business service being invoked (belonging to a certain RAG search tool in the third layer) is then activated, using its built-in multi-strategy fusion similarity algorithm to perform a precise search in the corresponding resource library to obtain the search results with the best match to the user's intent.
[0107] After the search is completed, the system returns the final result to the user in the corresponding data format of the business service called (such as structured data, resource access links, image files, etc.).
[0108] An exemplary embodiment of this disclosure also provides a business system access device, the structure of which is shown in FIG8, including: a global instruction receiving module 801, used to receive natural language instructions; a service matching module 802, used to filter target services that match the natural language instructions from an available service list, the available service list including multiple business services and description information corresponding to each business service; and a service routing module 803, used to initiate a call request to the target service based on a standard communication protocol.
[0109] The aforementioned apparatus can be integrated into a comprehensive platform supporting distributed deployment, with the corresponding functions implemented by a single server or server cluster on that platform. Regarding the apparatus in the above embodiments, the specific methods by which each module performs its operations have been described in detail in the embodiments related to the method, and will not be elaborated upon here.
[0110] An exemplary embodiment of this disclosure also provides a computer apparatus, including: a processor; a memory for storing processor-executable instructions; wherein the processor is configured to execute a business system access method provided by an embodiment of this disclosure.
[0111] An exemplary embodiment of this disclosure also provides a non-transitory computer-readable storage medium, wherein instructions in the storage medium, when executed by a processor of a computer device, enable the computer device to perform the business system access method provided by an embodiment of this disclosure.
[0112] Embodiments of this disclosure provide a business system access method, apparatus, computer device, and storage medium. The method involves receiving a natural language instruction, filtering a list of available services that matches the instruction, the list including multiple business services and corresponding descriptions; and then initiating a request to access the target service based on a standard communication protocol. By recognizing the user's natural language intent, routing business services based on a unified standard communication protocol, and managing different business services through a unified platform, this method solves the problems of low efficiency, inaccurate results, and high implementation costs in business system access, achieving a user-friendly, highly efficient, and accurate multimodal business access system.
[0113] By deeply integrating the MCP service architecture, large-scale model semantic understanding, and RAG retrieval technology, a unified intelligent access framework for business systems has been built, forming a complete closed loop of "service management - semantic understanding - retrieval execution".
[0114] In terms of service governance, it can manage a variety of heterogeneous business services in a unified manner, support dynamic service discovery and standardized invocation, and greatly reduce the complexity of service integration and maintenance.
[0115] In terms of search accuracy, relying on the multimodal RAG search architecture and business rule constraint mechanism, it integrates multiple strategies such as full matching, semantic matching, and business-specific numerical matching for weighted calculation, which improves the search accuracy from 50% to 90%, effectively adapting to the search needs of various types of business resources such as videos and flowcharts.
[0116] In terms of user experience, it supports direct operation of business systems using natural language, eliminating the need for users to master professional interface knowledge, reducing the learning cost by 70%, and successfully upgrading the interaction mode from "people finding functions" to "services finding people".
[0117] In terms of system performance, through browser pooling management and multi-level caching design, the business response time has been compressed from the traditional minutes to the seconds, which has greatly improved business processing efficiency.
[0118] It balances architectural innovation, efficient service governance, accurate retrieval, user-friendliness, and high system performance, comprehensively improving the intelligence level and practical application value of business systems.
[0119] Those skilled in the art will also understand that the various illustrative logical blocks and steps listed in the embodiments of this disclosure can be implemented by electronic hardware, computer software, or a combination of both. Whether such functionality is implemented in hardware or software depends on the specific application and the overall system design requirements. Those skilled in the art can implement the described functionality using various methods for each specific application, but such implementation should not be construed as exceeding the scope of protection of the embodiments of this disclosure.
[0120] Furthermore, the term “exemplary” is used herein to mean serving as an example, instance, or illustration. Any aspect or design described herein as “exemplary” is not necessarily to be construed as advantageous compared to other aspects or designs. Rather, the use of the term “exemplary” is intended to present the concept in a concrete manner. As used herein, the term “or” is intended to mean an inclusive “or” rather than an exclusive “or.” That is, unless otherwise specified or clear from the context, “X applies A or B” is intended to mean any of the natural inclusive arrangements. That is, “X applies A or B” satisfies any of the foregoing instances if X applies A; X applies B; or both X applies A and B. Additionally, unless otherwise specified or clear from the context to refer to the singular form, the articles “a” and “an” as used in this application and the appended claims are generally understood to mean “one or more.”
[0121] Similarly, although this disclosure has been shown and described with respect to one or more implementations, equivalent variations and modifications will occur to those skilled in the art upon reading and understanding this specification and the accompanying drawings. This disclosure includes all such modifications and variations and is limited only by the scope of the claims. In particular, with respect to the various functions performed by the components described above (e.g., elements, resources, etc.), unless otherwise indicated, the terminology used to describe such components is intended to correspond to any component (functionally equivalent) that performs the specific function of the described component, even if structurally not equivalent to the disclosed structure. Furthermore, although specific features of this disclosure may have been disclosed with respect to only one of several implementations, such features may be combined with one or more other features of other implementations, as may be desired and advantageous to any given or particular application. Moreover, with regard to the terms “comprising,” “owning,” “having,” “having,” or variations thereof as used in the detailed description or claims, such terms are intended to be inclusive in a manner similar to the term “including.”
[0122] Other embodiments of this disclosure will readily occur to those skilled in the art upon consideration of the specification and practice of the invention disclosed herein. This disclosure is intended to cover any variations, uses, or adaptations of this disclosure that follow the general principles of this disclosure and include common knowledge or customary techniques in the art not disclosed herein. The specification and examples are to be considered exemplary only, and the true scope and spirit of this disclosure are indicated by the following claims.
[0123] It should be understood that this disclosure is not limited to the precise structures described above and shown in the accompanying drawings, and various modifications and changes can be made without departing from its scope. The scope of this disclosure is limited only by the appended claims.
Claims
1. A method for accessing a business system, characterized in that, include: Receive natural language commands; The system filters the list of available services that match the natural language command. The list of available services includes multiple business services and descriptions of each business service. Based on a standard communication protocol, the system initiates a request to call the target service.
2. The business system access method according to claim 1, characterized in that, The method further includes: performing standardized registration of multiple business services on a registration management server, recording the name and description information of each business service; configuring a unified service discovery interface for multiple business services, and receiving call requests based on the standard communication protocol.
3. The business system access method according to claim 2, characterized in that, The step of filtering the target service matching the natural language instruction from the list of available services includes: obtaining the list of available services through the service discovery interface based on the standard communication protocol; parsing the natural language instruction to obtain the user intent; and determining the matching business service from the list of available services as the target service according to the user intent.
4. The business system access method according to claim 1, characterized in that, The step of initiating a call request to the target service based on a standard communication protocol includes: initiating the call request to the target service through a unified business execution interface based on the standard communication protocol.
5. The business system access method according to claim 1, characterized in that, The method further includes: obtaining a retrieval tool corresponding to the target service, wherein the retrieval tool includes at least any one or more of the following: a video surveillance retrieval tool for processing and retrieving video surveillance resources, a flowchart retrieval tool for processing and retrieving business approval flowcharts, a page effect diagram retrieval tool for processing and retrieving user interface and page design diagrams, and an application interface retrieval tool for processing and retrieving the operation interface of the actual application system; and calling the retrieval tool to perform multi-strategy fusion retrieval to obtain retrieval results.
6. The business system access method according to claim 5, characterized in that, The step of calling the retrieval tool to perform multi-strategy fusion retrieval and obtain retrieval results includes: calling the retrieval tool to obtain at least one original retrieval result; analyzing each of the original retrieval results according to the matching strategy, and taking the original retrieval result with the highest matching degree with the target service as the retrieval result. The matching strategy includes at least any one or more of the following: complete matching strategy, partial matching strategy, semantic matching strategy, fuzzy matching strategy, and numerical matching strategy for the target service.
7. The business system access method according to claim 1, characterized in that, The method further includes: obtaining the data type of the retrieval result of the target service request; and encapsulating the retrieval result generated in response to the call request according to the data type and feeding it back based on the standard communication protocol.
8. A business system access device, characterized in that, include: A global instruction receiving module is used to receive natural language instructions; a service matching module is used to filter target services that match the natural language instructions from an available service list, wherein the available service list includes multiple business services and description information corresponding to each business service. The service routing module is used to initiate a call request to the target service based on a standard communication protocol.
9. A computer device, characterized in that, include: processor; A memory for storing processor-executable instructions; wherein the processor is configured to execute the business system access method as described in any one of claims 1 to 7.
10. A non-transitory computer-readable storage medium, characterized in that, When the instructions in the storage medium are executed by the processor of the computer device, the computer device is able to perform the business system access method as described in any one of claims 1 to 7.
