Operation method and device of business system, storage medium and AI server
Dynamically generate JS instructions through AI servers, the problem of poor flexibility in the interactive instruction set of business systems is solved, efficient user intention recognition and response is achieved, maintenance costs are reduced, and user experience and system flexibility are improved.
Patent Information
- Application Number
- CN202411014258.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-26
- Publication Date
- 2024-11-26
AI Technical Summary
In the prior art, the interactive instruction set of the business system has poor flexibility and large maintenance workload, and cannot effectively respond to non-standard or novel user interaction scenarios, resulting in a decline in user experience and high maintenance costs.
Receive user conversation content through an AI server, use pre-trained large models to identify user intentions, and dynamically generate JS instructions to execute the user's expected page interaction effect, avoiding the need to manually update the instruction set.
It improves the flexibility and scalability of the system, reduces maintenance and upgrade costs, can accurately identify complex user intentions, and improves user experience and system intelligence level.
Smart Images

Figure CN119025083A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of business interaction, and in particular to an operating method, device, storage medium and AI server of a business system. Background Art
[0002] The traditional method of generating interactive instructions through the current dialogue has significant technical limitations. This method mainly relies on a limited set of pre-set instructions, which are fixed at the beginning of the design and trigger the corresponding execution actions through a simple keyword matching mechanism. However, with the continuous development of business systems and the increasing diversification of user needs, the fundamental problems of this approach have gradually been exposed.
[0003] First, the coverage of the preset instruction set is seriously insufficient. Since it is impossible to foresee all possible user interaction scenarios and changes in business needs in advance, the instruction set can only cover a part of common operations. When faced with non-standard or novel queries, the system often cannot provide effective responses, resulting in a decline in user experience.
[0004] Secondly, the lack of flexibility of traditional methods is also a major pain point. Since the instruction set is static, once the business logic or user needs change, these instructions need to be manually updated and maintained. This not only increases the workload and cost, but may also lead to response delays and errors. In addition, for complex or ambiguous user intentions, simple keyword matching mechanisms are often difficult to accurately capture, which leads to misunderstandings and misoperations. Summary of the invention
[0005] The embodiments of the present application provide an operation method, device, storage medium and AI server of a business system, which can solve the problems of poor flexibility and heavy maintenance workload of the instruction set of interactive instructions in the prior art. The technical solution is as follows:
[0006] In a first aspect, an embodiment of the present application provides an operating method of a business system, the method comprising:
[0007] The AI robot set on the current page displayed by the terminal device receives the conversation content input by the user;
[0008] Matching the view objects in the current page in the vector library according to the URL of the current page, and extracting the attribute information of each view object;
[0009] Using a pre-trained large model to identify the attribute information and the conversation content to obtain the user's conversation intention;
[0010] Generate JS instructions according to the conversation intention;
[0011] The JS instruction is sent to the terminal device so that the terminal device executes the JS instruction for the current page.
[0012] In a second aspect, an embodiment of the present application provides an operating device of a business system, the device comprising:
[0013] The transceiver unit is used to receive the dialogue content input by the user through the AI robot set on the current page displayed by the terminal device;
[0014] An extraction unit, used to match the view objects in the current page in the vector library according to the URL of the current page, and extract attribute information of each view object;
[0015] An identification unit, used to identify the attribute information and the conversation content using a pre-trained large model to obtain the user's conversation intention;
[0016] A generating unit, used for generating a JS instruction according to the conversation intention;
[0017] The transceiver unit is further used to send the JS instruction to the terminal device, so that the terminal device executes the JS instruction for the current page.
[0018] In a third aspect, an embodiment of the present application provides a computer storage medium, wherein the computer storage medium stores a plurality of instructions, wherein the instructions are suitable for being loaded by a processor and executing the above-mentioned method steps.
[0019] In a fourth aspect, an embodiment of the present application provides an AI server, which may include: a processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the above-mentioned method steps.
[0020] The beneficial effects brought about by the technical solutions provided by some embodiments of the present application include at least:
[0021] It can dynamically generate JS instructions for business systems. The content and logic of these instructions are not fixed by the program, but dynamically generated according to the actual needs of users and the status of the business system. This mechanism not only improves the flexibility and scalability of the system, but also significantly reduces the cost of maintenance and upgrades. Because when the business system is updated, there is no need to upgrade the functional modules to support the new instruction set, thus reducing unnecessary development workload and time cost. BRIEF DESCRIPTION OF THE DRAWINGS
[0022] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.
[0023] Figure 1 It is a schematic diagram of the network architecture provided by the embodiment of the present application;
[0024] Figure 2 It is a flowchart of an operation method of a business system provided in an embodiment of the present application;
[0025] Figure 3 It is a structural schematic diagram of an operating device of a business system provided by the present application;
[0026] Figure 4 This is a structural diagram of an AI server provided by this application. DETAILED DESCRIPTION
[0027] In order to make the objectives, technical solutions and advantages of the present application more clear, the embodiments of the present application will be further described in detail below with reference to the accompanying drawings.
[0028] It should be noted that the operating method of the business system provided in this application is generally executed by an AI server, and accordingly, the operating device of the business system is generally set in the AI server.
[0029] Figure 1 An exemplary system architecture that can be applied to the business system operation method or business system operation device of the present application is shown.
[0030] like Figure 1 As shown, the system architecture may include: an AI server 101 and a terminal device 102. The AI server 101 and the terminal device 102 may communicate via a network, and the network is used as a medium for providing communication links between the above-mentioned units. The network may include various types of wired communication links or wireless communication links, for example: a wired communication link includes an optical fiber, a twisted pair or a coaxial cable, etc., and a wireless communication link includes a Bluetooth communication link, a wireless fidelity (WIreless-FIdelity, Wi-Fi) communication link or a microwave communication link, etc.
[0031] It should be noted that the AI server 101 and the terminal device 102 can be hardware or software. When the AI server 101 and the terminal device 102 are hardware, they can be implemented as a distributed server cluster consisting of multiple servers, or as a single server. When the AI server 101 and the terminal device 102 are software, they can be implemented as multiple software or software modules (for example, to provide distributed services), or as a single software or software module, which is not specifically limited here.
[0032] Various communication client applications can be installed on the AI server of the present application, such as: video recording applications, video playback applications, voice interaction applications, search applications, instant messaging tools, email clients, social platform software, etc.
[0033] The AI server can be hardware or software. When the AI server is hardware, it can be various AI servers with display screens, including but not limited to smartphones, tablet computers, laptop computers, and desktop computers. When the AI server is software, it can be installed in the AI servers listed above. It can be implemented as multiple software or software modules (for example, to provide distributed services), or it can be implemented as a single software or software module, which is not specifically limited here.
[0034] When the AI server is hardware, a display device and a camera may also be installed on it. The display device may be any device that can realize the display function, and the camera is used to collect video streams; for example, the display device may be a cathode ray tube display (CR), a light-emitting diode display (LED), an electronic ink screen, a liquid crystal display (LCD), a plasma display panel (PDP), etc. Users can use the display device on the AI server to view displayed text, pictures, videos and other information.
[0035] It should be understood that Figure 1 The number of AI servers, networks, and terminal devices in the figure is only for illustration. Any number of AI servers, networks, and terminal devices may be used as required.
[0036] The following will be combined with the attached Figure 2 , the operation method of the business system provided in the embodiment of the present application is described in detail. Among them, the operation device of the business system in the embodiment of the present application can be Figure 1 The AI server shown.
[0037] See also Figure 2 , which is a flowchart of an operation method of a business system provided in an embodiment of the present application. Figure 2 As shown, the method of the embodiment of the present application may include the following steps:
[0038] S201. The AI robot set on the current page displayed by the terminal device receives the dialogue content input by the user.
[0039] The user inputs or clicks a link through the browser of the terminal device, and sends an HTTP GET request to the business system server to request the URL of the current page. After receiving the request, the business system server parses the page resources (HTML, CSS, JavaScript, etc.) to be returned according to the URL (Uniform Resource Locator) and sends it to the browser through an HTTP response. After receiving the response, the browser parses the HTML document, loads the CSS and JavaScript files, and renders the current page for the user to view.
[0040] The user enters the conversation content in the AI (Artificial Intelligence) robot component of the current page and triggers a submission action (such as clicking the send button). The front-end JavaScript code of the AI server captures the user input and sends the conversation content as a request body to the cloud AI server through AJAX, WebSocket, FetchAPI, etc. The request usually also contains the URL or other identification information of the current page so that the AI server can identify which page the user is interacting with. The cloud AI server receives the request from the terminal device and parses the conversation content entered by the user and the identification information of the current page.
[0041] S202: Match the view objects in the current page in the vector library according to the URL of the current page, and extract the attribute information of each view object.
[0042] Among them, the AI server extracts the URL of the current page from the request. The AI server uses the URL of the current page (or the page ID and other identifiers associated with it) as the query condition to retrieve the view object and its attribute information in the current page from the database. The database stores the structure of the page, the type and attributes of the view object (such as prompt information, type information, etc.) and other metadata. The AI server extracts the required view object attribute information from the database query results to prepare for the subsequent dialogue intent recognition.
[0043] In some embodiments of the present application, the attribute information of the view object includes Chinese prompt information, attribute name, element display name, value type, and element type.
[0044] When the AI server matches the view object and its attribute information in the current page from the vector library, it will parse this information, paying special attention to the Chinese prompt information, attribute name, element display name, value type, and element type.
[0045] Chinese prompt information: This is often used to show users additional instructions or help information about the view object on the user interface. In the processing of the AI server, it may be used to generate more humane dialogue responses or to verify whether the user input is as expected. Attribute name: This is a unique identifier for the view object attribute, which is used to reference and operate these attributes in internal logic. Element display name: This is the name displayed by the view object on the user interface, which may be used to reference or display to the user in the dialogue. Value type: This specifies the data type of the view object attribute value (such as string, number, Boolean value, etc.), which helps the AI server to perform type checking and conversion when parsing user input and generating JS instructions. Element type: This describes the type of view object (such as text box, button, drop-down list, etc.), which helps the AI server understand the user's operation intention on the page and generate corresponding JS instructions to simulate these operations.
[0046] In the process of identifying the user's conversational intent, the AI server will consider the above attribute information to provide a more accurate response. For example: If the user asks about the purpose of a view object or how to fill it out, the AI server can generate an explanatory conversation response based on the Chinese prompt information and the element display name. If the user tries to modify the value of the view object through a conversation, the AI server will check the value type to ensure that the data type entered by the user matches the expected value, and generate a JS instruction to update the view object value accordingly. If the user requests to perform a specific operation related to the view object (such as clicking a button), the AI server will generate a JS instruction to simulate the operation based on the element type.
[0047] In some embodiments of the present application, before S202, the following steps are further included:
[0048] The user is authenticated and the authentication result is passed.
[0049] Among them, the user accesses the business system through the browser on the terminal device and attempts to perform operations that require authentication (such as interacting with an AI robot). Before performing any operations related to the AI robot, the business system (possibly through the front-end code or the back-end API) will require the user to authenticate. Usually it involves submitting a username and password, using tokens (such as JWT), OAuth, single sign-on (SSO) and other mechanisms. After receiving the authentication request, the business system will execute the corresponding verification logic. This may include querying the database to verify the validity of the user's credentials, checking the validity of the token, or interacting with other authentication services. If the verification is successful, the business system will notify the AI server to proceed to the next steps; if the verification fails, the user request is rejected, and an error message may be returned or redirected to the login page.
[0050] S203: Use the pre-trained large model to identify the attribute information and the conversation content to obtain the user's conversation intention.
[0051] Among them, the processed attribute information and conversation content are passed as input to the pre-trained and possibly fine-tuned large model. The model will reason and analyze the input data based on the knowledge structure and patterns learned internally. The model will output one or more intent classification results, which are usually the understanding and summary of the user's conversation content. These intent classifications may correspond to specific operations or response types in the business system. Based on the intent classification results output by the model, the system may need to perform further confirmation or verification to ensure that the user's intention is correctly understood. This can be achieved by having an interactive conversation with the user, displaying a confirmation message, or checking business logic rules. Once the user's intention is confirmed, the system can generate corresponding JS instructions or other response content based on the intention and send it to the terminal device for execution or display. These instructions and response content will be dynamically generated based on the current state of the business system and the needs of the user. By collecting user feedback and the operation of the business system, the performance of conversation intent recognition can be continuously evaluated and optimized. Based on feedback and new data, the large model can be updated regularly to improve its accuracy and generalization ability. This can include using new training data, adjusting model parameters, or introducing new model architectures.
[0052] Through the above process, the pre-trained large model can efficiently identify the user's conversation intention and provide intelligent responses to the business system. This technology not only improves the efficiency and quality of interaction between users and business systems, but also reduces the cost of manual maintenance and upgrades.
[0053] S204: Generate JS instructions according to the dialogue intention.
[0054] Among them, the AI server formulates a corresponding response strategy based on the recognized user conversation intention and generates corresponding JavaScript (JS) instructions. These JS instructions are intended to be executed on the client to achieve the page interaction effect expected by the user. The AI server encodes the generated JS instructions into a string form so that it can be sent to the AI server through an HTTP response.
[0055] S205: Send the JS instruction to the terminal device, so that the terminal device executes the JS instruction for the current page.
[0056] The AI server uses the encoded JS instructions as part of the response body to construct an HTTP response. The response may also contain other metadata, such as status code, content type, etc. The AI server sends the constructed response to the terminal device that initiated the request (i.e., the user's browser) through an HTTP connection. After receiving the response, the browser of the terminal device parses the JS instructions and executes them. These instructions will be executed in the context of the current page to achieve user-desired operations such as updating page content, triggering animation effects, and calling page functions.
[0057] In some embodiments of the present application, the AI server scans each page of the business system, extracts the attribute information of the view objects of each page, and stores the extracted attribute information in a database for learning and training of the large model.
[0058] Among them, the AI server first starts a web crawler (Web Crawler), which can traverse various pages of the business system, including sub-pages navigated through links. The crawler should be able to handle different types of page structures (such as HTML, JavaScript dynamically generated content, etc.). For each page visited, the crawler will parse its DOM (Document Object Model) structure to identify view objects in the page (such as input boxes, buttons, drop-down lists, etc.). This usually involves parsing HTML tags and attributes, as well as JavaScript code that may be executed (such as simulating browser behavior through tools such as Selenium). The crawler extracts the attribute information of each view object, including Chinese prompt information, attribute name, element display name, value type, element type, etc. This information may be stored in the attributes of the HTML tag, or it may be dynamically generated by JavaScript code. The extracted attribute information will be formatted into an appropriate data structure (such as JSON or XML) and stored in the database. The database design should support efficient query and indexing for subsequent rapid retrieval of this information.
[0059] Choose a suitable database system (such as MySQL, PostgreSQL, MongoDB, etc.) according to business requirements and data volume. Design database tables to store the attribute information of view objects. Each table may contain fields such as page ID, view object ID, Chinese prompt information, attribute name, element display name, value type, element type, etc. Create indexes for database tables to improve query performance. For example, you can create indexes for page ID and view object ID to quickly retrieve attribute information for a specific page or view object. Verify data before storing it in the database to ensure data integrity and consistency.
[0060] Retrieve the extracted view object attribute information from the database and perform necessary preprocessing (such as data cleaning, normalization, feature extraction, etc.) to meet the requirements of the AI model. Select appropriate AI models (such as deep learning models, natural language processing models, etc.) for training. These models can use the attribute information of view objects to learn the patterns and rules of user interaction with business systems. Use the validation set to evaluate the trained model, and adjust and optimize the model based on the evaluation results to improve its accuracy and performance. Deploy the trained model to the AI server for real-time reasoning and prediction in actual business scenarios.
[0061] The user interacts with the AI robot on the business system page and enters the conversation content. The AI server receives the conversation content entered by the user, and uses the deployed AI model to identify the intent in combination with the view object attribute information stored in the database. Based on the identified user intent, the AI server generates the corresponding JS instructions or other response content and sends it to the AI server for execution or display.
[0062] Through the above process, the AI server can automatically learn the page structure and user interaction mode of the business system, thereby improving the level of intelligent interaction with users. At the same time, this automated method also reduces the burden of manual intervention and improves the scalability and maintainability of the system.
[0063] In some embodiments of the present application, after the AI server recognizes the user's conversation intention, it does not directly operate the business system, but uses the JS instruction generation engine to dynamically build JS instructions. These JS instructions are designed for the front-end page or specific components of the business system, and are intended to simulate the user's interactive behavior or trigger specific business logic. According to the user's intention and the current state of the business system, the JS instruction generation engine will build a JS code snippet that conforms to the business logic. These code snippets may include operations such as positioning, attribute modification, and event triggering of page elements. When generating JS instructions, the AI server will ensure the security of the instructions to avoid executing malicious code or leaking sensitive information. At the same time, the compatibility of the instructions will also be considered to ensure that they can be correctly executed on different versions of browsers and AI servers. The generated JS instructions will be encapsulated into a format suitable for transmission (such as a JSON string) and sent to the AI server over the network.
[0064] After receiving the JS instructions sent by the AI server, the terminal device will use the built-in JS execution engine to execute these instructions. The JS execution engine is a part of the browser or terminal device that is responsible for parsing and executing JavaScript code. The terminal device first parses the received JS instructions and restores them to executable JavaScript code. Before executing the instructions, the terminal device will ensure that the front-end page of the business system has been loaded and the relevant DOM elements are ready. The JS execution engine will operate on the front-end page or component of the business system according to the requirements in the instructions. This may include modifying the attribute values of page elements, triggering button click events, calling specific JavaScript functions, etc. After executing the instructions, the terminal device may feedback the execution results or changes in page status to the AI server or user. This helps the AI server evaluate the execution effect of the instructions and make adjustments as needed.
[0065] The embodiments of the present application specifically include the following beneficial effects:
[0066] By scanning and learning each page of the business system and automatically extracting the attribute information of the view object, the solution achieves a high degree of adaptability to business system updates. When the business system changes, there is no need to manually update the preset instructions in the functional module, thus avoiding the problem of requiring a lot of manpower to maintain and upgrade instructions due to business system updates in traditional methods. This automation and adaptability significantly improves the flexibility and response speed of the system.
[0067] Relying on a large model to judge the user's conversational intent is more effective and accurate than the traditional text keyword matching method. The large model can understand more complex contexts and semantics. Even in the face of typos and word order errors in user input, it can accurately identify the user's true intent, thereby avoiding incorrect responses caused by misunderstandings. This intelligent intent recognition capability greatly improves the user experience and the intelligence level of the system.
[0068] Based on intent recognition and extraction of key information, the solution can dynamically generate JS instructions for business systems. The content and logic of these instructions are not fixed by the program, but are dynamically generated according to the actual needs of users and the status of the business system. This mechanism not only improves the flexibility and scalability of the system, but also significantly reduces the cost of maintenance and upgrades. Because when the business system is updated, there is no need to upgrade the functional modules to support the new instruction set, thus reducing unnecessary development workload and time costs.
[0069] The following is an embodiment of the device of the present application, which can be used to execute the embodiment of the method of the present application. For details not disclosed in the embodiment of the device of the present application, please refer to the embodiment of the method of the present application.
[0070] See also Figure 3 , which shows a schematic diagram of the structure of an operating device of a business system provided by an exemplary embodiment of the present application, hereinafter referred to as device 3. The device 3 can be implemented as all or part of the AI server through software, hardware or a combination of both. The device 3 includes: a transceiver unit 301, an extraction unit 302, an identification unit 303, and a generation unit 304.
[0071] The transceiver unit 301 is used to receive the dialogue content input by the user through the AI robot set on the current page displayed by the terminal device;
[0072] An extraction unit 302 is used to match the view objects in the current page in the vector library according to the URL of the current page, and extract attribute information of each view object;
[0073] The recognition unit 303 is used to use a pre-trained large model to recognize the attribute information and the conversation content to obtain the user's conversation intention;
[0074] A generating unit 304, configured to generate a JS instruction according to the conversation intention;
[0075] The transceiver unit 301 is further configured to send the JS instruction to the terminal device, so that the terminal device executes the JS instruction for the current page.
[0076] In one or more possible embodiments, it further includes:
[0077] The scanning unit is used to scan each page of the business system, extract the attribute information of the view object of each page, and store the extracted attribute information in a database for learning and training of the large model.
[0078] In one or more possible embodiments, it further includes:
[0079] The verification unit is used to verify the identity of the user and the verification result is pass.
[0080] In one or more possible embodiments, the attribute information of the view object includes: Chinese prompt information, attribute name, element display name, value type, and element type.
[0081] In one or more possible embodiments, sending the JS instruction to the terminal device includes:
[0082] The generated JS instruction is encoded into a character string, and an HTTP response carrying the character string is sent to the terminal device.
[0083] In one or more possible embodiments, the AI server generates JS instructions using a JS instruction generation engine, and the terminal device executes the JS instructions using a JS execution engine.
[0084] In one or more possible embodiments, the AI robot set on the current page displayed by the terminal device receives the dialogue content input by the user, including:
[0085] A request body sent from a terminal device is received via HTTP, wherein the request body carries the URL of the current page and the conversation content input by the user collected by the AI robot.
[0086] It should be noted that, when the device 3 provided in the above embodiment executes the method for operating the business system, only the division of the above functional modules is used as an example. In actual applications, the above functions can be assigned to different functional modules as needed, that is, the internal structure of the device is divided into different functional modules to complete all or part of the above functions. In addition, the operating device of the business system provided in the above embodiment and the embodiment of the method for operating the business system belong to the same concept, and the implementation process thereof is detailed in the method embodiment, which will not be repeated here.
[0087] The serial numbers of the above-mentioned embodiments of the present application are for description only and do not represent the advantages or disadvantages of the embodiments.
[0088] The present application also provides a computer storage medium, which can store multiple instructions, and the instructions are suitable for being loaded and executed by a processor as described above. Figure 2The method steps of the embodiment shown in the figure can be found in the specific implementation process. Figure 2 The specific description of the illustrated embodiment will not be repeated here.
[0089] The present application also provides a computer program product, which stores at least one instruction, and the at least one instruction is loaded and executed by the processor to implement the operation method of the business system described in the above embodiments.
[0090] See also Figure 4 , which is a schematic diagram of the structure of an AI server provided in the embodiment of the present application. Figure 4 As shown, the AI server 400 may include: at least one processor 401 , at least one network interface 404 , a user interface 403 , a memory 405 , and at least one communication bus 402 .
[0091] The communication bus 402 is used to realize the connection and communication between these components.
[0092] The user interface 403 may include a display screen (Display) and a camera (Camera), and the optional user interface 403 may also include a standard wired interface and a wireless interface.
[0093] The network interface 404 may optionally include a standard wired interface or a wireless interface (such as a WI-FI interface).
[0094] Among them, the processor 401 may include one or more processing cores. The processor 401 uses various interfaces and lines to connect the various parts of the entire AI server 400, and executes various functions and processes data of the AI server 400 by running or executing instructions, programs, code sets or instruction sets stored in the memory 405, and calling data stored in the memory 405. Optionally, the processor 401 can be implemented in at least one hardware form of digital signal processing (Digital Signal Processing, DSP), field programmable gate array (Field-Programmable Gate Array, FPGA), and programmable logic array (Programmable Logic Array, PLA). The processor 401 can integrate one or a combination of a central processing unit (Central Processing Unit, CPU), a graphics processing unit (Graphics Processing Unit, GPU), and a modem. Among them, the CPU mainly processes the operating system, user interface, and application programs; the GPU is responsible for rendering and drawing the content to be displayed on the display screen; and the modem is used to process wireless communications. It can be understood that the above-mentioned modem may not be integrated into the processor 401, but may be implemented separately through a chip.
[0095] Among them, the memory 405 may include a random access memory (Random Access Memory, RAM) and may also include a read-only memory (Read-Only Memory). Optionally, the memory 405 includes a non-transitory computer-readable storage medium. The memory 405 can be used to store instructions, programs, codes, code sets or instruction sets. The memory 405 may include a program storage area and a data storage area, wherein the program storage area may store instructions for implementing an operating system, instructions for at least one function (such as a touch function, a sound playback function, an image playback function, etc.), instructions for implementing the above-mentioned method embodiments, etc.; the data storage area may store data involved in the above-mentioned method embodiments, etc. The memory 405 may optionally be at least one storage device located away from the aforementioned processor 401. As Figure 4 As shown, the memory 405 as a computer storage medium may include an operating system, a network communication module, a user interface module, and an application program.
[0096] exist Figure 4In the AI server 400 shown in FIG. 1 , the user interface 403 is mainly used to provide an input interface for the user and obtain the data input by the user; and the processor 401 can be used to call the application stored in the memory 405 and specifically execute the following steps: Figure 2 The specific process can be referred to Figure 2 As shown, no further description is given here.
[0097] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiments can be implemented by instructing related hardware through a computer program, and the program can be stored in a computer-readable storage medium, and when the program is executed, it can include the processes of the embodiments of the above-mentioned methods. The storage medium can be a disk, an optical disk, a read-only storage memory, or a random access memory, etc.
[0098] The above disclosure is only the preferred embodiment of the present application, which certainly cannot be used to limit the scope of rights of the present application. Therefore, equivalent changes made according to the claims of the present application are still within the scope covered by the present application.
Claims
1. A method for operating a business system, characterized in that: include: The AI robot set on the current page displayed by the terminal device receives the conversation content input by the user; Matching the view objects in the current page in the vector library according to the URL of the current page, and extracting the attribute information of each view object; Using a pre-trained large model to identify the attribute information and the conversation content to obtain the user's conversation intention; Generate JS instructions according to the conversation intention; The JS instruction is sent to the terminal device so that the terminal device executes the JS instruction for the current page.
2. The method according to claim 1, characterized in that: Before the AI robot in the current page receives the dialogue content input by the user, it also includes: Scan each page of the business system, extract the attribute information of the view object of each page, and store the extracted attribute information in a database for learning and training of the large model.
3. The method according to claim 1 or 2, characterized in that: Also includes: The identity of the user is verified and the verification result is passed.
4. The method according to claim 3, characterized in that The attribute information of the view object includes: Chinese prompt information, attribute name, element display name, value type, and element type.
5. The method according to claim 1, 2 or 4, characterized in that: The sending the JS instruction to the terminal device includes: The generated JS instruction is encoded into a character string, and an HTTP response carrying the character string is sent to the terminal device.
6. The method according to claim 5, characterized in that The AI server generates JS instructions using a JS instruction generation engine, and the terminal device executes the JS instructions using a JS execution engine.
7. The method according to claim 1 or 2 or 4 or 6, characterized in that: The AI robot set on the current page displayed by the terminal device receives the dialogue content input by the user, including: A request body sent from a terminal device is received via HTTP, wherein the request body carries the URL of the current page and the conversation content input by the user collected by the AI robot.
8. An operating device for a business system, characterized in that: include: The transceiver unit is used to receive the dialogue content input by the user through the AI robot set on the current page displayed by the terminal device; An extraction unit, used to match the view objects in the current page in the vector library according to the URL of the current page, and extract attribute information of each view object; An identification unit, used to identify the attribute information and the conversation content using a pre-trained large model to obtain the user's conversation intention; A generating unit, used for generating a JS instruction according to the conversation intention; The transceiver unit is further used to send the JS instruction to the terminal device, so that the terminal device executes the JS instruction for the current page.
9. A computer storage medium, characterized in that The computer storage medium stores a plurality of instructions, and the instructions are suitable for being loaded by a processor and executing the method steps according to any one of claims 1 to 7.
10. An AI server, characterized in that: include: A processor and a memory; wherein the memory stores a computer program, and the computer program is suitable for being loaded by the processor and executing the method steps as claimed in any one of claims 1 to 7.
