Tabular data processing method, device, medium and program product

Abstract

The present invention provides a table data processing method that can be applied to the field of artificial intelligence. The table data processing method includes: inputting table data and associated processing requests input by an object via an interactive interface into a preset model, generating a tool call request corresponding to the processing request, and sending the tool call request to a task processing engine; inputting tool information returned by the task processing engine into the preset model to obtain multiple target tools, including reading tools and processing tools; when it is determined that the task processing engine has completed reading the table data using the reading tool, inputting the reading result and functional information of the processing tool into the preset model to obtain a processing task corresponding to the processing request, and sending the processing task to the task processing engine; and in response to receiving an execution result from the task processing engine, displaying the execution result on the interactive interface. The present invention also provides an electronic device, a storage medium, and a program product.

Description

Technical Field

[0001] The present invention relates to the field of artificial intelligence, and in particular to a table data processing method, device, medium and program product. Background Art

[0002] When processing tabular data, we often encounter tables containing millions of rows and complex multi-table relational analysis scenarios. Related technologies utilize large models to achieve intelligent processing of tabular data. After learning tool functions, large models can perform various data operations, basic operations, and complex analysis.

[0003] In the process of realizing the concept of the present invention, there are at least the following problems in the related art: due to the wide variety of data operations, large models are less efficient in learning tool functions and are difficult to meet the diverse needs of users, resulting in low processing efficiency. Summary of the Invention

[0004] In view of the above problems, the present invention provides a table data processing method, apparatus, device, medium and program product.

[0005] According to a first aspect of the present invention, a table data processing method is provided, comprising: inputting table data and associated processing requests input by an object via an interactive interface into a preset model, generating a tool call requirement corresponding to the above processing request, and sending the above tool call requirement to a task processing engine; inputting the tool information returned by the above task processing engine into the above preset model to obtain a plurality of target tools, and the above plurality of target tools include a reading tool and a processing tool; when it is determined that the above task processing engine completes reading the above table data through the above reading tool, inputting the reading result and the functional information of the above processing tool into the above preset model to obtain a processing task corresponding to the above processing request, and sending the above processing task to the above task processing engine; in response to receiving the execution result from the above task processing engine, displaying the above execution result on the above interactive interface.

[0006] The second aspect of the present invention provides a table data processing device, including: a demand generation module, which is used to input the table data and the associated processing request input by the object through the interactive interface into a preset model, generate a tool call demand corresponding to the above processing request, and send the above tool call demand to the task processing engine; a tool determination module, which is used to input the tool information returned by the above task processing engine into the above preset model to obtain multiple target tools, and the above multiple target tools include reading tools and processing tools; a task generation module, which is used to input the reading result and the functional information of the above processing tool into the above preset model when it is determined that the above task processing engine completes reading the above table data through the above reading tool, obtain a processing task corresponding to the above processing request, and send the above processing task to the above task processing engine; a result display module, which is used to display the above execution result on the above interactive interface in response to receiving the execution result from the above task processing engine.

[0007] A third aspect of the present invention provides an electronic device, comprising: one or more processors; a memory for storing one or more computer programs, wherein the one or more processors execute the one or more computer programs to implement the steps of the above method.

[0008] The fourth aspect of the present invention further provides a computer-readable storage medium having a computer program or instructions stored thereon, which implements the steps of the above method when the computer program or instructions are executed by a processor.

[0009] The fifth aspect of the present invention further provides a computer program product, comprising a computer program or instructions, which implement the steps of the above method when executed by a processor.

[0010] According to an embodiment of the present invention, by inputting tabular data and processing requests into a preset model to generate tool call requirements, interacting with the task processing engine to determine the target tool, and then combining the read results and the functional information of the processing tool to generate and execute the processing task, this process leverages the collaborative mechanism of the model and engine to achieve the dynamic generation of tool call requirements and processing tasks. This overcomes the limitations of manual intervention in tool matching and task generation in existing data processing, supports full process automation from data reading to task execution, and improves the accuracy of tool matching and processing efficiency. Furthermore, the real-time presentation of results through an interactive interface solves the problems of cumbersome and inefficient processing procedures. BRIEF DESCRIPTION OF THE DRAWINGS

[0011] The above contents and other objects, features and advantages of the present invention will become more apparent through the following description of the embodiments of the present invention with reference to the accompanying drawings.

[0012] Figure 1A diagram illustrating an application scenario of a table data processing method, apparatus, device, medium, and program product according to an embodiment of the present invention is shown.

[0013] Figure 2 A flowchart of a table data processing method according to an embodiment of the present invention is shown.

[0014] Figure 3 A schematic diagram of the architecture of a table data processing method according to an embodiment of the present invention is shown.

[0015] Figure 4 A software architecture diagram of a table data processing method according to an embodiment of the present invention is shown.

[0016] Figure 5 A schematic diagram of an interactive interface of a table data processing method according to an embodiment of the present invention is shown.

[0017] Figure 6 An interactive flow chart of a table data processing method according to an embodiment of the present invention is shown.

[0018] Figure 7 A structural block diagram of a table data processing apparatus according to an embodiment of the present invention is shown.

[0019] Figure 8 A block diagram of an electronic device suitable for implementing a table data processing method according to an embodiment of the present invention is shown. DETAILED DESCRIPTION

[0020] Hereinafter, embodiments of the present invention will be described with reference to the accompanying drawings. However, it should be understood that these descriptions are exemplary only and are not intended to limit the scope of the present invention. In the following detailed description, for ease of explanation, many specific details are set forth to provide a comprehensive understanding of embodiments of the present invention. However, it is apparent that one or more embodiments may also be implemented without these specific details. In addition, in the following description, descriptions of known structures and technologies are omitted to avoid unnecessary confusion of the concept of the present invention.

[0021] The terms used herein are only for describing specific embodiments and are not intended to limit the present invention. The terms "comprise", "include", etc. used herein indicate the presence of the features, steps, operations and / or components, but do not exclude the presence or addition of one or more other features, steps, operations or components.

[0022] All terms used herein (including technical and scientific terms) have the meanings commonly understood by those skilled in the art unless otherwise defined. It should be noted that the terms used herein should be interpreted as having a meaning consistent with the context of this specification and should not be interpreted in an idealized or overly rigid manner.

[0023] When expressions such as "at least one of A, B, and C, etc." are used, they should generally be interpreted in accordance with the meaning commonly understood by those skilled in the art (for example, "a system having at least one of A, B, and C" should include but is not limited to a system having A alone, B alone, C alone, A and B, A and C, B and C, and / or A, B, C, etc.).

[0024] In the technical solution of the present invention, the data involved (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, storage, use, processing, transmission, provision, disclosure and application of the relevant data comply with relevant laws, regulations and standards, take necessary confidentiality measures, do not violate public order and good morals, and provide corresponding operation entrances for users to choose to authorize or refuse.

[0025] An embodiment of the present invention provides a method for processing table data, including: inputting table data and associated processing requests input by an object through an interactive interface into a preset model, generating a tool call requirement corresponding to the processing request, and sending the tool call requirement to a task processing engine; inputting tool information returned by the task processing engine into the preset model to obtain multiple target tools, and the multiple target tools include a reading tool and a processing tool; when it is determined that the task processing engine completes reading the table data through the reading tool, inputting the reading result and the functional information of the processing tool into the preset model to obtain a processing task corresponding to the processing request, and sending the processing task to the task processing engine; in response to receiving the execution result from the task processing engine, displaying the execution result on the interactive interface.

[0026] Figure 1 A diagram illustrating an application scenario of a table data processing method, apparatus, device, medium, and program product according to an embodiment of the present invention is shown.

[0027] like Figure 1 As shown, the application scenario 100 according to this embodiment may include a first terminal device 101, a second terminal device 102, a third terminal device 103, and a server 105. A network 104 is used as a medium for providing a communication link between the first terminal device 101, the second terminal device 102, the third terminal device 103, and the server 105. The network 104 may include various connection types, such as wired or wireless communication links or optical fiber cables.

[0028] A user may use a first terminal device 101, a second terminal device 102, or a third terminal device 103 to interact with a server 105 via a network 104 to receive or send messages, etc. Various communication client applications may be installed on the first terminal device 101, the second terminal device 102, or the third terminal device 103, such as shopping applications, web browser applications, search applications, instant messaging tools, email clients, social platform software, etc. (for example only).

[0029] The first terminal device 101 , the second terminal device 102 , and the third terminal device 103 may be various electronic devices having display screens and supporting web browsing, including but not limited to smart phones, tablet computers, laptop computers, desktop computers, and the like.

[0030] The server 105 may be a server that provides various services, such as a background management server (for example only) that supports websites browsed by users using the first terminal device 101, the second terminal device 102, and the third terminal device 103. The background management server may analyze and process received data such as user requests, and feed back processing results (e.g., web pages, information, or data obtained or generated based on user requests) to the terminal devices.

[0031] It should be noted that the table data processing method provided in the embodiment of the present invention can generally be executed by the server 105. Accordingly, the table data processing apparatus provided in the embodiment of the present invention can generally be set in the server 105. The table data processing method provided in the embodiment of the present invention can also be executed by a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105. Accordingly, the table data processing apparatus provided in the embodiment of the present invention can also be set in a server or server cluster that is different from the server 105 and can communicate with the first terminal device 101, the second terminal device 102, the third terminal device 103 and / or the server 105.

[0032] It should be understood that Figure 1 The number of the first terminal device, the second terminal device, the third terminal device, the network and the server is only . According to the implementation requirements, there can be any number of the first terminal device, the second terminal device, the third terminal device, the network and the server.

[0033] The following will be based on Figure 1 The scene described by Figures 2 to 6 The table data processing method of the embodiment is described in detail.

[0034] Figure 2 A flowchart of a table data processing method according to an embodiment of the present invention is shown.

[0035] like Figure 2 As shown, this embodiment includes operations S210 to S240.

[0036] In operation S210 , the tabular data and the associated processing request input by the object through the interactive interface are input into a preset model, a tool calling requirement corresponding to the processing request is generated, and the tool calling requirement is sent to the task processing engine.

[0037] In operation S220 , the tool information returned by the task processing engine is input into a preset model to obtain a plurality of target tools, where the plurality of target tools include a reading tool and a processing tool.

[0038] In operation S230, when it is determined that the task processing engine completes reading the table data through the reading tool, the reading result and the functional information of the processing tool are input into the preset model to obtain a processing task corresponding to the processing request, and the processing task is sent to the task processing engine.

[0039] In operation S240 , in response to receiving the execution result from the task processing engine, the execution result is displayed on the interactive interface.

[0040] According to embodiments of the present invention, with the acceleration of digital transformation, businesses and organizations face increasingly complex data processing needs. This is especially true when processing tabular data, which often requires combining multiple tools and algorithms to complete specific tasks. By combining artificial intelligence models (such as the Large Language Model (LLM)) with tool invocation and establishing a standardized interaction framework based on the Model Context Protocol (MCP), the system automatically selects and invokes the appropriate tool based on the user's processing request, achieving efficient processing of tabular data.

[0041] In the interactive interface, after users enter tabular data and processing requests, the input data is first preprocessed, including format verification and data cleansing, and converted into a standardized format that the pre-set model can understand. MCP, acting as an intermediate protocol layer, converts the input data into a standardized contextual format that the pre-set model (such as LLM) can understand. The pre-set model then performs in-depth semantic analysis of the processing request, combining the structural and content characteristics of the tabular data to intelligently generate tool call requirements. MCP then encapsulates these tool call requirements into structured messages and sends them to the task processing engine.

[0042] Upon receiving a tool request, the task processing engine dynamically queries the available tool library, accurately identifies the tool set that matches the request, and returns tool information such as the tool name, function description, and input and output formats to the preset model. This information is then re-entered into the preset model through the MCP. The preset model then selects the optimal target tool from the candidate tools based on the core objectives of the processing request and the specific characteristics of the table data. This includes two key components: reading tools and processing tools.

[0043] Through MCP, precise instructions are sent to the task processing engine, prioritizing the use of the reading tool to parse the table data. The reading tool accurately extracts the target data from the table data according to the format specifications and data rules defined by MCP, and returns the structured reading results to the preset model. After confirming that the data reading operation is complete, the reading results and processing tool function information are input into the preset model through MCP. Based on the unified context specified by MCP, the preset model comprehensively analyzes data characteristics and tool capabilities to generate specific processing tasks. The processing tasks are then organized into an MCP-defined executable instruction pipeline and sent to the task processing engine.

[0044] After the task processing engine completes all processing tasks, it returns the final execution results through MCP. At this point, the results can be further processed, such as format conversion and result verification, and then visualized through the interactive interface. The display format intelligently adapts based on the type of processing request and the characteristics of the execution result, supporting various MCP-compatible presentation methods such as table views, statistical charts, and text summaries, ensuring that users can intuitively understand the processing results.

[0045] By inputting tabular data and processing requests into a preset model to generate tool call requirements, interacting with the task processing engine to determine the target tool, and then combining the read results and the functional information of the processing tool to generate and execute the processing task, this process leverages the collaborative mechanism of the model and engine to achieve dynamic generation of tool call requirements and processing tasks. This breaks through the limitations of manual intervention in tool matching and task generation in existing data processing, supports full process automation from data reading to task execution, and improves the accuracy of tool matching and processing efficiency. At the same time, the real-time presentation of results through the interactive interface solves the problems of cumbersome and inefficient processing procedures.

[0046] Figure 3 A schematic diagram of the architecture of a table data processing method according to an embodiment of the present invention is shown.

[0047] like Figure 3As shown, in the process framework for collaboratively processing tabular data 303 based on a preset model 301 and an MCP server 302, the user first uploads tabular data 303 to the preset model 301 as a data input source. Simultaneously, the user's requirements, questions, or instructions constitute processing requests 304, which are input into the preset model 301 along with the tabular data 303.

[0048] The preset model 301 processes the input table data 303 and processing request 304 by virtue of its semantic understanding and other capabilities, and then passes the processing task 305 to the task processing engine 306 in the MCP server 302 for execution, thereby realizing intelligent processing and analysis of the table data 303.

[0049] According to an embodiment of the present invention, the tool information returned by the task processing engine is input into a preset model to obtain multiple target tools, including: sending a tool call request to the task processing engine to select multiple candidate tools from the tool registration library of the task processing engine; inputting the functional information of each candidate tool returned by the task processing engine into the preset model to determine multiple target tools that match the processing request from the multiple candidate tools.

[0050] In the data processing flow, once a preset model generates a tool call requirement, the tool call requirement is standardized and encapsulated through MCP and sent to the task processing engine. This process ensures the standardization and consistency of the requirement description, allowing the task processing engine to accurately understand the processing intent.

[0051] After receiving a tool request formatted in MCP, the task processing engine screens candidate tools from its own tool registry based on the metadata specifications defined by MCP. MCP provides a unified description framework for each tool in the tool registry (including functional signatures, input and output formats, performance metrics, etc.), enabling the engine to quickly and standardizedly match the characteristics of the tool request with the tool metadata, thereby accurately identifying the set of tools that may meet the requirements.

[0052] MCP retrieves the functional information of each candidate tool returned by the task processing engine, all in the contextual format specified by MCP. MCP's standardized description ensures comparability and consistency across tools, laying the foundation for subsequent intelligent selection. This standardized functional information is fully transferred to the pre-defined model, where specific algorithms within the pre-defined model perform in-depth analysis based on the structured data provided by MCP.

[0053] The pre-configured model uses the standardized evaluation dimensions established by the MCP (such as functional fit, applicability, and processing efficiency) to conduct a multi-dimensional comparison of each candidate tool's capabilities against the processing request. The MCP serves as a benchmark for evaluation criteria, ensuring that comparisons between different tools are conducted on the same basis. By parsing contextual data in the MCP format, the pre-configured model accurately understands each tool's capability boundaries and application scenarios.

[0054] After intelligent computation using the pre-defined model, the MCP outputs the final target tools. These tools are selected based on objective evaluation within the MCP specification to ensure their precise functionality meets the processing requirements. The MCP also formats the selected tools into standard call instructions to support subsequent task processing.

[0055] Through the calculation and judgment of the preset model, multiple target tools that match the processing request are determined from multiple candidate tools. These target tools can more accurately meet the requirements of the processing request in terms of functionality, providing more targeted support for subsequent tool calls and task processing, ensuring that the system can complete the corresponding tasks efficiently and accurately.

[0056] According to an embodiment of the present invention, the functional information of each candidate tool returned by the task processing engine is input into a preset model to determine multiple target tools that match the processing request from multiple candidate tools, including: extracting industry characteristic information from the processing request and table data, wherein the industry characteristic information includes business keywords and data structure characteristics; through the semantic analysis function, determining the target knowledge base corresponding to the industry characteristic information from the preset industry knowledge base; according to the configuration requirements of the target knowledge base, analyzing the functional information of each candidate tool to extract multiple target tools that match the processing request from the candidate tools.

[0057] We extract industry-specific information from processing requests and tabular data. This process leverages natural language processing and data structure analysis to accurately identify business keywords (e.g., "risk assessment" and "compliance audit" in the financial sector) and data structure characteristics (e.g., the type, format, or relationships of specific fields in tabular data). This industry-specific information accurately reflects the industry attributes and data characteristics of the processing request.

[0058] Utilizing semantic analysis, we deeply match the extracted industry characteristics with the pre-set industry knowledge base. Semantic analysis deeply understands the connotations of business keywords and the industry significance of data structure characteristics, thereby locating the corresponding target knowledge base within the pre-set industry knowledge base.

[0059] For example, if the extracted business keywords involve medical image analysis and the data structure characteristics are medical image format, the semantic analysis function will determine the target knowledge base related to medical image processing, which stores the industry's professional knowledge, standard specifications and common processing patterns.

[0060] Based on the target knowledge base's configuration requirements, we conduct a detailed analysis of the functional information of each candidate tool. These requirements clearly define the functional standards and performance indicators required to handle tasks in this industry scenario. By comparing these requirements, we extract key elements that match the processing requirements from the candidate tools' functional descriptions, technical specifications, and applicable scenarios.

[0061] This targeted analysis allows us to filter out candidate tools to identify those whose functionality precisely meets industry requirements and can effectively handle the current task. This ensures that the tool selection is both industry-specific and meets the specific requirements of the request, laying a solid foundation for subsequent tool deployment.

[0062] In addition, to optimize the efficiency of calling preset models and external tools, a three-level caching mechanism and asynchronous call architecture can be adopted. A vector cache library is deployed on the model side to convert frequently used tool call parameters (such as data reading paths and processing logic templates) into vector indexes and store them. When calling, similarity matching is used for rapid retrieval, reducing the time spent on repeated inference. The external tool interface is encapsulated as a microservice cluster and configured with an adaptive thread pool to dynamically adjust the concurrency based on the load. When consecutive call requests of the same type are detected, batch mode is automatically triggered, merging multiple requests into a single call to reduce communication overhead. Through these optimizations, the response time of tool calls can be effectively reduced.

[0063] According to an embodiment of the present invention, the table processing method further includes: sending the tool identifications of multiple target tools to the task processing engine, so that the task processing engine performs structured reading of the table data using a reading tool in the multiple target tools based on the initial storage address of the table data.

[0064] The tool identifiers of multiple target tools are encapsulated according to the MCP to form a standardized tool call request. This tool call request not only contains the unique identifier of the target tool but also includes contextual metadata for processing the request, such as the session identifier, timestamp, and call parameters, to ensure the task processing engine accurately understands the call intent. This request is then sent to the task processing engine via a secure and reliable communication channel.

[0065] After receiving a tool call request, the task processing engine first parses the request to determine the initial storage address of the table data. This initial storage address can be a local file path, a network storage location, or a database connection string, and it specifies the physical storage location of the table data. Next, the task processing engine selects a tool with table read functionality from the target tools and instantiates it.

[0066] When the Read tool is called, it accesses the table data based on the initial storage address. It then uses the Read function to load the file and identify the header structure based on the number of header rows pre-entered by the user. For multi-layered headers, the Read function automatically creates a hierarchical index object, converting nested headers into a hierarchical multi-level index. The Read tool further analyzes the data type of the header, identifying different types of fields, such as text, numbers, and dates.

[0067] During the reading process, the reading tool records the basic structural information of the table data, including the total number of rows and columns, and whether there are special cases such as merged cells. For merged cells, the reading tool applies specific algorithms to ensure data integrity and accuracy. For complex table layouts, such as nested tables or irregular structures, the reading tool uses a recursive parsing strategy to extract structural information layer by layer, ultimately converting the entire table data into a structured data object, including the complete header hierarchy and data content, providing a clear and standardized data foundation for subsequent data processing and analysis.

[0068] According to an embodiment of the present invention, the reading results and functional information of the processing tool are input into a preset model to obtain a processing task corresponding to the processing request, including: generating a prompt word template based on the initial storage address of the table data, the reading results, the processing request, the preset storage address of the execution result, and the preset code execution format; calling the preset model based on the prompt word template to generate code for processing the table data to form a processing task.

[0069] A prompt word template is dynamically constructed based on the table data's initial storage address, read results, processing request, the preset storage address of the execution results, and the preset code execution format. During the construction process, key features of the table data are first extracted from the read results. For example, the table data's subject can be determined by analyzing the table header and data content, while the data range is defined based on the number of rows and columns in the table data and the data's value range. This information is organized into a brief description of the table data and incorporated into the prompt word template.

[0070] For example, if the initial storage address is ". / excel / regional gross domestic product.xlsx", the read result shows that the table header is ['Region', '2022 gross domestic product (100 million)', 'Industry', '2023 gross domestic product (100 million)']. Based on this, the table description is generated: "In the ". / excel / regional gross domestic product.xlsx" table, the table header is ['Region', '2022 gross domestic product (100 million)', 'Industry', '2023 gross domestic product (100 million)']".

[0071] The processing request will be parsed into specific operation instructions. For example, if the processing request is "Which three industries appear most frequently in the table? Write a new row", it will be converted into a clear operation description and used as the third line of prompt words.

[0072] The preset storage address of the execution results will be formatted as a path representation that meets the storage system's requirements, such as "Save to a new table in the '. / excel' folder." The preset code execution format will be converted into syntax and structure constraints for the generated code, such as "Output only computer programming language (Python code), output as "python."

[0073] Based on the generated prompt word template, the preset model is called. When called, the prompt word template is passed as input to the preset model. The preset model will deeply analyze the table description, operation instructions, storage requirements and other information in the prompt word based on its own semantic understanding ability.

[0074] For simple processing requests, the preset model quickly matches the appropriate data processing functions and logic, generating concise and clear code snippets. For complex processing requests, such as those involving multi-table joins or data pivoting, the preset model constructs more complex code logic to ensure that the full functionality of the processing request can be achieved.

[0075] During code generation, the pre-set model fully considers the data type and structure of the tabular data, selecting the most appropriate functions and methods to handle different types of data. For example, mathematical functions are prioritized for numeric data, while string processing functions are used for textual data. The generated code adheres to good coding standards, with clear comments and a reasonable code structure to improve readability and maintainability.

[0076] If the tabular data contains unusual data structures or formats, the pre-set model will add corresponding processing logic to the code to ensure that the code can correctly handle various exceptions. The resulting code is encapsulated as a complete processing task, including the entire process of data reading, processing, and result storage. It can be directly called and executed by the execution engine, thus realizing automated processing of tabular data.

[0077] According to an embodiment of the present invention, the table data processing method further includes: performing static syntax verification on the code using a preset model; and when it is determined that the verification passes, sending the generated processing task to the task processing engine.

[0078] Automatically launch the preset model to perform static syntax verification on the generated code. During the verification process, the preset model carefully checks each component of the code according to the grammatical rules of the Python language. It checks whether the key words in the code, such as if, else, for, etc., are spelled correctly; checks whether the statement structure is complete, for example, whether each function definition contains a colon and correct indentation, and whether each code block has a reasonable hierarchy; and checks whether symbols such as brackets and quotation marks appear in pairs and are used correctly. For expressions in the code, the preset model checks whether the use of operators complies with grammatical rules and whether variables have been defined before use.

[0079] The pre-configured model also performs a more in-depth syntax analysis, checking for problems such as unclosed statements, incorrect indentation, and mismatched brackets. It simulates the code's execution flow, checking that variable scopes are appropriate and that function calls pass the correct number and types of parameters. For complex code logic, the pre-configured model analyzes control flow statements, such as loops and conditionals, to ensure they are logically correct and do not result in infinite loops or undefined behavior.

[0080] After confirming that the code has passed the static syntax check, the generated processing task is sent to the task processing engine. The processing task is encapsulated into a task package containing code, execution parameters, and context information. This task package follows the interface specifications of the task processing engine and contains necessary metadata, such as task identification, the source of the processing request, and execution priority. The task package is sent to the task processing engine via a reliable communication mechanism to ensure that the task is not lost or damaged during transmission. After receiving the task package, the task processing engine prepares the execution environment based on the information contained therein and executes the code according to the predetermined process to complete the task of processing the table data. By accurately identifying syntax errors and logical risks before code execution, the exception rate during code execution is reduced, providing an efficient and reliable execution foundation for subsequent table data processing.

[0081] According to an embodiment of the present invention, the table data processing method also includes: sending the processing task to the sandbox environment in the task processing engine for execution; wherein, during the execution process, the processing task is subjected to an environmental check, and the environmental check includes determining at least one of whether the dependent package environment matches, whether the third-party library is available, and whether the environment configuration is compatible; and when it is determined that the environment check passes, the execution result is output.

[0082] The task is sent to the Python sandbox environment within the task processing engine for execution. This sandbox uses lightweight container technology for resource isolation, limiting memory usage to 512MB and processor time quota to 30 seconds. Before task execution, the sandbox automatically launches a dependency resolver, builds a virtual environment based on the files in the task, and generates a hash to lock dependency versions to ensure environmental consistency.

[0083] The environment verification phase employs a multi-layered verification strategy: First, static analysis tools scan code import statements to extract a list of required third-party libraries; then, modules are called to dynamically check installed package versions within the sandbox environment; and finally, smoke tests are executed to verify the proper functioning of key functions. For missing dependencies, an error signature extraction mechanism is triggered. By parsing command error output, regular expressions are used to match missing package names and versions, generating a structured error log.

[0084] If the environment verification is confirmed to have passed, the execution results are output. This verification mechanism significantly improves the success rate of code execution and system stability by dynamically monitoring resource usage and continuously iterating and optimizing.

[0085] According to an embodiment of the present invention, the table data processing method also includes: terminating the execution of the processing task when it is determined that the result obtained by the environmental check meets the preset conditions; in response to receiving the structured error log generated when the task processing engine performs the environmental check, optimizing and reconstructing the code through a preset model.

[0086] When the environment verification results meet pre-set conditions (such as dependency package mismatches, unavailable third-party libraries, or environment configuration conflicts), the task termination mechanism is immediately triggered. The monitoring module within the sandbox environment detects abnormal signals during the environment verification process in real time. If it detects missing critical dependencies or incompatible environment configurations, it sends a termination instruction to the task executor through inter-process communication, while preserving the current execution context to prevent data loss. The termination process adopts a gradual strategy, first pausing the task thread execution and then releasing allocated system resources to ensure a clean sandbox environment.

[0087] After receiving the structured error log generated by the task processing engine during the environment check, the log content is input into the preset model for optimization and reconstruction analysis. The semantic parsing module of the preset model converts the structured information contained in the structured error log, such as missing dependency names, version constraints, and error types, into executable optimization instructions. The preset model uses abstract syntax tree technology to traverse the code structure and locate the code segments that need modification based on the error characteristics. For example, this can include adding statements for missing third-party libraries or adjusting function call parameters to adapt to the environment configuration.

[0088] During the refactoring analysis process, the pre-set model combines the code's functional logic and environmental requirements to generate multiple optimization solutions and evaluate their feasibility. After the refactoring is complete, the pre-set model performs static syntax validation and semantic verification on the optimized code to ensure that the refactoring does not introduce new errors. The optimized code, along with the updated dependency configuration files, forms a new processing task and is sent to the task processing engine's sandbox environment for verification until the environment passes the verification.

[0089] This automated error response and code refactoring mechanism can quickly resolve environmental incompatibilities and reduce the rate of code execution failures caused by environmental issues. Furthermore, the feedback loop of structured error logs enables the pre-set model to continuously learn and optimize refactoring strategies. As processing tasks accumulate, the model's efficiency in resolving complex environmental issues significantly improves, achieving an evolution from passive response to active optimization and ensuring the stability and compatibility of processing tasks across diverse environments.

[0090] According to an embodiment of the present invention, the table data processing method also includes: comparing the table data and the execution results through the reading rules of the document processing library in the task processing engine to obtain a comparison result; based on the difference content in the comparison result, marking and performing data verification in the execution result to generate a verification report, wherein the data verification includes at least one of format verification, range verification and logical relationship verification; the verification report and the execution results marked with the difference content are visually displayed through an interactive interface.

[0091] The document processing library in the task processing engine is called to parse the original table data and the target file after code execution according to the preset reading rules. With the help of the underlying interface, the cell data of the source and target files are traversed row by row and column by column, and the differences in values, text, format, and other dimensions are quickly located through hash value comparison technology. For numerical data, the comparison is performed within a tolerance range of ±10%, and those outside this range are marked as outliers. For text data, the cosine similarity algorithm is used to calculate the similarity cosθ. The specific calculation method is shown in Formula (1):

[0092]

[0093] Where A and B represent the eigenvalues ​​of two text vectors respectively, n represents the total number of eigenvalues ​​of the text vectors, and i represents the loop variable of the summation operation.

[0094] When the similarity falls below a preset value (80%), a difference flag is triggered, and a regular expression is used to verify that the character set contains no illegal characters. The format verification module automatically extracts style attributes such as fonts, borders, and alignment, generating a structured format descriptor for element-by-element comparison.

[0095] After obtaining the comparison results, visual annotations are made in the execution result document based on the differences. Cells with numeric differences are highlighted with a red fill and a yellow border; text differences are underlined in green; and cells with format differences are outlined in blue, with the specific differences noted in the annotations.

[0096] A multi-dimensional data validation process is also initiated: the range validation module performs logical analysis based on the numerical ranges preset by business rules (e.g., GDP data must be positive). Logical relationship validation verifies that row and column calculation relationships conform to pre-set rules (e.g., whether the total column equals the sum of the sub-items) by constructing a data dependency graph. During the validation process, a verification report is generated in real time, storing detailed information on any discrepancies, including the location, type (numeric / text / format), comparison between the original and target values, and the validation rule basis.

[0097] Verification reports and annotated execution results are visually displayed through an interactive interface. The interface uses a split-screen mode, with the original table data displayed on the left and the annotated execution results on the right. Difference cells are dynamically highlighted and displayed in real time. Users can add feedback using the interface's built-in annotation system, which automatically synchronizes this feedback to the code optimization module, forming a closed-loop verification-feedback-optimization process, ensuring the accuracy of execution results and a close match with user needs.

[0098] According to an embodiment of the present invention, the table data processing method also includes: in response to receiving the execution result from the task processing engine, analyzing the execution result based on the prompt word template to obtain an analysis result; when it is determined that the analysis result indicates that other tools need to be called to optimize the execution result, sending a call request for the optimization tool to the task processing engine, so that the task processing engine calls the corresponding optimization tool to optimize the execution result according to the call request.

[0099] After receiving the execution result returned by the task processing engine, the original prompt word template is automatically loaded, and the execution result is semantically aligned with the processing request, table description, and other information in the prompt word template. The content structure of the execution result is parsed through natural language processing technology. For example, for table operation tasks, the data statistics, screening results, or format change records in the result are extracted and bidirectionally compared with the user's original requirements in the prompt word template (such as "statistical high-frequency industries" and "write new rows"). Pre-set evaluation rules are applied during the analysis process, such as checking whether the numerical calculation results are within a reasonable error range, whether the text processing results conform to the semantic logic, and whether the format adjustment meets the storage requirements specified by the template, so as to generate structured analysis results.

[0100] When analysis indicates that the execution results do not fully meet user needs (for example, the statistical results omit some data, the output format does not meet the template requirements, or there are logical errors), the specific dimensions that need to be optimized are automatically identified, and a call request for the optimization tool is generated based on the prompt word template. The call request will clearly define the optimization goal (such as "supplementing missing data rows," "correcting numerical calculation logic," and "adjusting cell format"), input and output standards, and contextual constraints, such as the storage path and data type requirements in the original prompt word template. The call request is encapsulated into a standardized message according to the MCP, including the optimization tool's functional label (such as "data cleaning tool" and "format conversion tool") and a parameter list.

[0101] After the task processing engine receives the call request from the optimization tool, it will match the optimization tool with the corresponding function from the tool registration library. For example, it will call the data completion tool to handle missing values, or start the format conversion tool to convert the result file to a specified format. During the processing, the optimization tool will inherit the context information of the original prompt word template to ensure that the optimization logic is consistent with the user's initial requirements. For example, if the original prompt word template requires the results to be saved in a specific folder, the optimization tool will automatically use this storage path. After the optimization is completed, the new execution result will be returned for cyclic verification again until the analysis result is confirmed to fully meet the requirements of the prompt word template, forming a closed-loop processing process to ensure the ultimate realization of user needs.

[0102] Based on the analysis results, the optimization tool is dynamically called to optimize the execution results, forming a closed-loop optimization mechanism, realizing automatic optimization and iterative processing of the execution results, and improving the quality and adaptability of data processing.

[0103] Figure 4 A software architecture diagram of a table data processing method according to an embodiment of the present invention is shown.

[0104] like Figure 4 As shown, the layered architecture for processing tabular data includes interaction logic between each layer. Application layer 401 serves as the user interaction portal. When a user has an operation request, they initiate a command to service layer 402 via an application programming interface (API). For example, when a user performs file processing or data reasoning operations on the front-end interface, these requests are communicated by application layer 401.

[0105] The service layer 402 plays a core scheduling role. After receiving requests from the application layer 401, it interacts with the data layer 403 and the reasoning layer 404, respectively, based on the type of request. If a file-related operation (such as reading, writing, or storing) is involved, the service layer 402 sends a file operation request to the data layer 403 and obtains file data and metadata as needed. If an inference task is required, the service layer 402 sends an inference request to the reasoning layer 404. After the inference is complete, the service layer 402 receives the inference results and transmits the integrated feedback back to the application layer 401 to ensure the smooth operation of the business process.

[0106] The data layer 403 focuses on the storage and management of file data. It responds to file operation requests from the service layer 402 and provides or saves file data and metadata (such as file format and creation time). The inference layer 404 focuses on inference computing. It applies algorithms and models to inference requests from the service layer 402 and outputs inference results, providing intelligent analysis capabilities for the business. Each layer collaborates and divides work to support the system in completing data processing and inference tasks.

[0107] According to an embodiment of the present invention, the table processing method further includes: when it is determined that the analysis result indicates that the execution result meets the processing request, displaying a corresponding natural language response on the interactive interface according to the execution result.

[0108] After confirming that the execution result fully complies with the processing request, the execution result undergoes semantic abstraction and natural language conversion. First, the data structure of the execution result is parsed. For example, if the result is tabular data, key statistical values, screening results, or operation conclusions (such as "The three most frequently appearing industries in the table are manufacturing, information technology, and finance") are extracted. If the result is a file operation, the operation status (such as "Successfully saved to the specified folder") and key parameters (such as the file name and storage path) are extracted.

[0109] Dynamically generate natural language responses based on the user's query intent in the prompt word template. For example, if the processing request is for data statistics, the natural language response will include specific values ​​and trend analysis (such as "Province A will have the highest GDP in 2023, with a value of 12.8 trillion yuan, a 5.2% increase from 2022"); if it is a format adjustment, the natural language response will explain the format change details (such as "The table header font has been changed to Microsoft YaHei, font size 12, bold"). The content of the natural language response will be semantically enhanced based on the business context of the table data (such as industry attributes and data meaning) to ensure that the natural language expression conforms to the professional habits of the user's field.

[0110] The generated natural language response is visualized through the interactive interface. The interactive interface can adopt a card-style layout, displaying the original table data or a thumbnail of the execution result on the left, and a natural language summary in rich text on the right. Key data is highlighted with labels (such as green font for growth data and red font for outliers). For execution results containing complex logic (such as data analysis after multi-condition filtering), the natural language response will be accompanied by a diagram of the logical deduction process to help users understand the basis for the result generation. Generating natural language responses in the interactive interface provides natural language visual feedback of the processing results, improves the user-friendly human-computer interaction, and facilitates user understanding of the execution results.

[0111] According to an embodiment of the present invention, the table data processing method also includes: using the semantic understanding function in the preset model to analyze the data content in the execution results to obtain the data change trend; based on the data change trend, generating a visual statistical chart for display in the interactive interface.

[0112] The pre-set model's semantic understanding function performs in-depth analysis of the data content in the execution results. First, lexical analysis identifies business keywords (such as "date," "sales," and "age distribution") in the table header fields. The model then infers the data type (numeric, categorical, or time series) based on the context. For numeric data, the pre-set model calculates the interquartile range and detects outliers to determine the data distribution. If the data conforms to a normal distribution and has minimal influence from extreme values ​​(such as average employee age), the pre-set model automatically marks it as suitable for mean analysis. If significant outliers are present (such as revenue data), the model prioritizes the median. For categorical fields (such as "product category"), the pre-set model counts the frequency of each category and identifies the mode. It also calculates the standard deviation and variance, generating natural language explanations based on the business scenario (such as "The standard deviation of sales across regions is 253,000 yuan, indicating significant regional variations in sales performance").

[0113] Based on the above analysis results, the optimal visualization chart type is automatically selected. For time series data (such as "Monthly Sales"), a line chart with a trendline is generated, with the X-axis formatted using time labels (such as "YYYY-MM") and the Y-axis labeled with numerical ranges. For comparative categorical data (such as "Employee Satisfaction by Department"), a bar chart with error bars is generated, using industry-standard color schemes (such as blue for financial data). For percentage analysis (such as "Revenue Contribution by Product Line"), a doughnut-shaped pie chart with percentage labels is generated, highlighting components that contribute more than 20%. Data enhancement techniques are automatically applied during chart generation: moving average curves are added to line charts, data labels are added to bar charts, and effects are added to pie charts to highlight key categories.

[0114] The generated visualization charts are displayed through responsive components within the interactive interface. The interface features a split-screen layout, with a collapsible data panel on the left, allowing users to switch between different data subsets. The right side houses the visualization area, which includes the main chart and a thumbnail navigation bar. Users can quickly locate the time period of interest by dragging the thumbnail. Hovering the mouse cursor displays the specific value and year-on-year change rate. Chart titles and descriptive text, such as "Revenue Distribution by Product Line, Q1 2023 (Red Areas Indicate Year-on-Year Declines of More Than 15%)," are automatically generated based on the analysis results to help users quickly understand the data.

[0115] Figure 5 A schematic diagram of an interactive interface of a table data processing method according to an embodiment of the present invention is shown.

[0116] like Figure 5 As shown, the operating interface of the local table tool 501 shows the data display and interaction process. The "Urban GDP Data Table" on the left presents the GDP, population, type, year and other data of cities in multiple provinces in a tabular form, intuitively displaying the basic information. The table dialogue 502 area on the right is the interactive area, with an upload location 5021 for uploading table data at the top and a dialogue content 5022 between the tool assistant and the user at the bottom. The function of the tool assistant itself is to analyze and process the uploaded table data, and lists the operation steps, including reading files, checking data, extracting specific city data, processing formats, sorting and checking, etc., clearly showing the processing logic of the table data. This helps users understand how tools collaborate to complete data operations and realizes the combination of data display and interactive analysis.

[0117] By using the preset model semantic understanding function to analyze the execution result data, identify the change trend and generate a visual statistical chart display, it realizes the intelligent analysis and visual presentation of data trends, helping users to intuitively grasp the data dynamics and improve the efficiency of data interpretation and interactive experience.

[0118] Figure 6 An interactive flow chart of a table data processing method according to an embodiment of the present invention is shown.

[0119] like Figure 6 As shown, the process for processing tabular data based on a preset model includes operations S1 through S10. After the user initiates an operation, they upload the processing request and the tabular data to the preset model. The preset model first reads the data structure and then sends a tool call request to the task processing engine. Based on the tool call request, the task processing engine calls the target tool to read the tabular data and returns the read result to the preset model.

[0120] The preset model constructs a prompt word template based on the read results, then calls its own code to generate and manipulate the table data. This code is then packaged as a processing task and sent to the task processing engine for execution. Once the task processing engine completes, it automatically initiates a result verification mechanism, comparing the results with the original data and generating a verification report containing the data differences. Both the results and the verification report are then fed back to the preset model.

[0121] Preset models conduct in-depth analysis of execution results, extracting key indicators and trends. This process, combined with table content reading and multidimensional data analysis, generates visualizations. Ultimately, these results are presented as interactive reports through a user-friendly interface, forming a complete closed loop from data input and processing to verification and analysis. This process, through layered collaboration and intelligent feedback, significantly improves the automation of table data processing and the accuracy of results.

[0122] Based on the above table data processing method, the present invention also provides a table data processing device, which will be described in detail below with reference to the figures.

[0123] Figure 7 A structural block diagram of a table data processing apparatus according to an embodiment of the present invention is shown.

[0124] like Figure 7 As shown, the table data processing device 700 of this embodiment includes a demand generation module 710 , a tool determination module 720 , a task generation module 730 and a result presentation module 740 .

[0125] The demand generation module 710 is configured to input the tabular data and associated processing requests entered by the subject via the interactive interface into a preset model, generate a tool call request corresponding to the processing request, and send the tool call request to the task processing engine. In one embodiment, the demand generation module 710 can be configured to perform operation S210 described above, and will not be further described here.

[0126] The tool determination module 720 is configured to input the tool information returned by the task processing engine into the preset model to obtain a plurality of target tools, wherein the plurality of target tools include a reading tool and a processing tool. In one embodiment, the tool determination module 720 can be configured to perform the operation S220 described above, which will not be further described here.

[0127] The task generation module 730 is configured to, upon determining that the task processing engine has completed reading the table data using the reading tool, input the reading result and the functional information of the processing tool into the preset model, obtain a processing task corresponding to the processing request, and send the processing task to the task processing engine. In one embodiment, the task generation module 730 can be configured to perform operation S230 described above, which will not be further described here.

[0128] The result display module 740 is used to display the execution result on the interactive interface in response to receiving the execution result from the task processing engine. In one embodiment, the result display module 740 can be used to perform the operation S240 described above, which will not be repeated here.

[0129] According to an embodiment of the present invention, any multiple modules among the demand generation module 710, tool determination module 720, task generation module 730, and result presentation module 740 may be combined into a single module, or any one of these modules may be split into multiple modules. Alternatively, at least part of the functionality of one or more of these modules may be combined with at least part of the functionality of other modules and implemented in a single module. According to an embodiment of the present invention, at least one of the demand generation module 710, tool determination module 720, task generation module 730, and result presentation module 740 may be at least partially implemented as a hardware circuit, such as a field programmable gate array (FPGA), a programmable logic array (PLA), a system on a chip, a system on a substrate, a system on a package, an application-specific integrated circuit (ASIC), or may be implemented in hardware or firmware by any other reasonable means of circuit integration or packaging, or may be implemented in any one of the three implementation methods of software, hardware, and firmware, or in any appropriate combination of any of these. Alternatively, at least one of the demand generation module 710 , the tool determination module 720 , the task generation module 730 and the result presentation module 740 may be at least partially implemented as a computer program module, which may perform corresponding functions when executed.

[0130] It should be noted that the table data processing device part in the embodiment of the present invention corresponds to the table data processing method part in the embodiment of the present invention. The description of the table data processing device part specifically refers to the table data processing method part and will not be repeated here.

[0131] Figure 8 A block diagram of an electronic device suitable for implementing a table data processing method according to an embodiment of the present invention is shown.

[0132] like Figure 8As shown, an electronic device 800 according to an embodiment of the present invention includes a processor 801, which can perform various appropriate actions and processes based on programs stored in a read-only memory (ROM) 802 or programs loaded from a storage unit 808 into a random access memory (RAM) 803. The processor 801 may include, for example, a general-purpose microprocessor (e.g., a CPU), an instruction set processor and / or related chipsets and / or a special-purpose microprocessor (e.g., an application-specific integrated circuit (ASIC)), etc. The processor 801 may also include onboard memory for caching purposes. The processor 801 may include a single processing unit or multiple processing units for performing different actions of the method flow according to an embodiment of the present invention.

[0133] The RAM 803 stores various programs and data required for the operation of the electronic device 800. The processor 801, ROM 802, and RAM 803 are connected to each other via a bus 804. The processor 801 executes the programs in the ROM 802 and / or RAM 803 to perform the various operations of the method flow according to the embodiment of the present invention. It should be noted that the programs may also be stored in one or more memories other than the ROM 802 and RAM 803. The processor 801 may also execute the programs stored in one or more memories to perform the various operations of the method flow according to the embodiment of the present invention.

[0134] According to an embodiment of the present invention, electronic device 800 may further include an input / output (I / O) interface 805, which is also connected to bus 804. Electronic device 800 may also include one or more of the following components connected to I / O interface 805: an input section 806 including a keyboard, mouse, etc.; an output section 807 including devices such as a cathode ray tube (CRT), liquid crystal display (LCD), and speakers; a storage section 808 including a hard disk; and a communication section 809 including a network interface card such as a LAN card or modem. Communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to I / O interface 805 as needed. Removable media 811, such as a magnetic disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed in drive 810 as needed, so that computer programs read from the removable media can be installed into storage section 808 as needed.

[0135] The present invention also provides a computer-readable storage medium, which may be included in the device / apparatus / system described in the above embodiments, or may exist independently and not incorporated into the device / apparatus / system. The computer-readable storage medium carries one or more programs, which, when executed, implement the method according to the embodiments of the present invention.

[0136] According to an embodiment of the present invention, a computer-readable storage medium may be a non-volatile computer-readable storage medium, and may include, for example, but not limited to: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a portable compact disk read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. For example, according to an embodiment of the present invention, a computer-readable storage medium may include the ROM 802 and / or RAM 803 described above, and / or one or more memories other than ROM 802 and RAM 803.

[0137] An embodiment of the present invention further includes a computer program product comprising a computer program containing program code for executing the method shown in the flowchart. When the computer program product is executed in a computer system, the program code is used to cause the computer system to implement the table data processing method provided by the embodiment of the present invention.

[0138] The computer program executes the above functions defined in the system / device of the embodiment of the present invention when executed by the processor 801. According to the embodiment of the present invention, the system, device, module, unit, etc. described above can be implemented by a computer program module.

[0139] In one embodiment, the computer program may be stored on a tangible storage medium such as an optical storage device or a magnetic storage device. In another embodiment, the computer program may be transmitted and distributed in the form of a signal on a network medium, downloaded and installed via the communication portion 809, and / or installed from a removable medium 811. The program code contained in the computer program may be transmitted using any appropriate network medium, including but not limited to wireless, wired, or any suitable combination thereof.

[0140] In such an embodiment, the computer program can be downloaded and installed from a network via the communication section 809 and / or installed from a removable medium 811. When the computer program is executed by the processor 801, the above-described functions defined in the system of the embodiment of the present invention are performed. According to the embodiment of the present invention, the systems, devices, means, modules, units, etc. described above can be implemented by computer program modules.

[0141] According to an embodiment of the present invention, the program code for executing the computer program provided by the embodiment of the present invention can be written in any combination of one or more programming languages. Specifically, these computer programs can be implemented using high-level procedural and / or object-oriented programming languages, and / or assembly / machine languages. Programming languages ​​include, but are not limited to, languages ​​such as Java, C++, Python, "C" or similar programming languages. The program code can be executed entirely on the user computing device, partially on the user device, partially on a remote computing device, or entirely on a remote computing device or server. In the case of a remote computing device, the remote computing device can be connected to the user computing device through any type of network, including a local area network (LAN) or a wide area network (WAN), or can be connected to an external computing device (for example, using an Internet service provider to connect via the Internet).

[0142] The flowcharts and block diagrams in the accompanying drawings illustrate the possible implementation architecture, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. In this regard, each box in the flowchart or block diagram can represent a module, program segment, or a part of code, and the above-mentioned module, program segment, or a part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the box can also occur in an order different from that marked in the accompanying drawings. For example, two boxes represented in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented with a dedicated hardware-based system that performs the specified function or operation, or can be implemented with a combination of dedicated hardware and computer instructions.

[0143] It will be understood by those skilled in the art that the features described in the various embodiments of the present invention may be combined and / or coupled in various ways, even if such combinations or couplings are not explicitly described in the present invention. In particular, the features described in the various embodiments of the present invention may be combined and / or coupled in various ways without departing from the spirit and teachings of the present invention. All such combinations and / or couplings fall within the scope of the present invention.

[0144] The above describes embodiments of the present invention. However, these embodiments are for illustrative purposes only and are not intended to limit the scope of the present invention. Although each embodiment has been described separately above, this does not mean that the measures in each embodiment cannot be advantageously used in combination. Without departing from the scope of the present invention, those skilled in the art may make various substitutions and modifications, which should all fall within the scope of the present invention.

Claims

1. A table data processing method, characterized in that: The method comprises: Inputting the tabular data and the associated processing request input by the object through the interactive interface into a preset model, generating a tool call requirement corresponding to the processing request, and sending the tool call requirement to the task processing engine; Inputting the tool information returned by the task processing engine into the preset model to obtain a plurality of target tools, wherein the plurality of target tools include a reading tool and a processing tool; When it is determined that the task processing engine has completed reading the table data through the reading tool, the reading result and the function information of the processing tool are input into the preset model to obtain a processing task corresponding to the processing request, and the processing task is sent to the task processing engine; In response to receiving the execution result from the task processing engine, the execution result is displayed on the interactive interface.

2. The method according to claim 1, characterized in that The tool information returned by the task processing engine is input into the preset model to obtain a plurality of target tools, including: Sending the tool call requirement to the task processing engine to select a plurality of candidate tools from a tool registry of the task processing engine; The function information of each candidate tool returned by the task processing engine is input into the preset model to determine a plurality of target tools matching the processing request from the plurality of candidate tools.

3. The method according to claim 2, characterized in that Inputting the function information of each candidate tool returned by the task processing engine into the preset model to determine a plurality of target tools matching the processing request from the plurality of candidate tools includes: Extracting industry characteristic information from the processing request and the table data, wherein the industry characteristic information includes business keywords and data structure characteristics; Determine the target knowledge base corresponding to the industry characteristic information from the preset industry knowledge base through the semantic analysis function; According to the configuration requirements of the target knowledge base, the functional information of each of the candidate tools is analyzed to extract a plurality of target tools that match the processing request from the candidate tools.

4. The method according to claim 1, wherein The method further comprises: The tool identifications of the multiple target tools are sent to the task processing engine, so that the task processing engine performs structured reading of the table data using a reading tool among the multiple target tools based on an initial storage address of the table data.

5. The method according to claim 1, characterized in that The step of inputting the reading result and the functional information of the processing tool into the preset model to obtain a processing task corresponding to the processing request includes: Generating a prompt word template based on the initial storage address of the table data, the read result, the processing request, the preset storage address of the execution result, and the preset code execution format; The preset model is called based on the prompt word template to generate code for processing the table data to form the processing task.

6. The method according to claim 5, characterized in that The method further comprises: Performing static syntax verification on the code using the preset model; When it is determined that the verification is passed, the generated processing task is sent to the task processing engine.

7. The method according to claim 5, characterized in that The method further comprises: Sending the processing task to the sandbox environment in the task processing engine for execution; During the execution process, an environment check is performed on the processing task, and the environment check includes determining at least one of whether the dependent package environment matches, whether the third-party library is available, and whether the environment configuration is compatible; If it is determined that the environment check has passed, the execution result is output.

8. The method according to claim 7, characterized in that The method further comprises: When it is determined that the result of the environmental verification satisfies the preset conditions, terminating the execution of the processing task; In response to receiving a structured error log generated when the task processing engine performs an environment check, the code is optimized and reconstructed and analyzed using the preset model.

9. The method according to claim 1, characterized in that The method further comprises: Comparing the table data with the execution result using the reading rules of the document processing library in the task processing engine to obtain a comparison result; Based on the differences in the comparison results, marking the execution results and performing data verification to generate a verification report, wherein the data verification includes at least one of format verification, range verification, and logical relationship verification; The verification report and the execution results marked with the difference content are visually displayed through the interactive interface.

10. The method according to claim 5, characterized in that The method further comprises: In response to receiving the execution result from the task processing engine, analyzing the execution result based on the prompt word template to obtain an analysis result; When it is determined that the analysis result indicates that other tools need to be called to optimize the execution result, a call request for the optimization tool is sent to the task processing engine so that the task processing engine calls the corresponding optimization tool to optimize the execution result according to the call request.

11. The method according to claim 10, characterized in that The method further comprises: When it is determined that the analysis result indicates that the execution result complies with the processing request, a corresponding natural language response is displayed on the interactive interface based on the execution result.

12. The method according to claim 1, characterized in that The method further comprises: Utilizing the semantic understanding function in the preset model, the data content in the execution result is analyzed to obtain the data change trend; A visual statistical chart is generated based on the data change trend to be displayed in the interactive interface.

13. An electronic device comprising: one or more processors; a memory for storing one or more computer programs, It is characterized in that the one or more processors execute the one or more computer programs to implement the steps of the method according to any one of claims 1 to 12.

14. A computer-readable storage medium having a computer program or instruction stored thereon, characterized in that: When the computer program or instructions are executed by a processor, the steps of the method according to any one of claims 1 to 12 are implemented.

15. A computer program product, characterized in that The invention comprises a computer program, which implements the method according to any one of claims 1 to 12 when being executed by a processor.

Images