Table processing method, electronic device, storage medium, and computer program product

Abstract

Embodiments of the present application provide a table processing method, an electronic device, a storage medium and a computer program product. The table processing method comprises: processing a target table file by a generative model to obtain standard data table data corresponding to the target table file; storing the standard data table data in a target storage unit, and establishing an external table of the standard data table data in an analysis database corresponding to the target storage unit; processing by the generative model based on metadata of the external table to generate a data query instruction; and executing the data query instruction to determine a data analysis result for the target table file based on the external table. The present scheme can realize intelligent table analysis.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular to a table processing method, electronic device, storage medium, and computer program product. Background Technology

[0002] Intelligent questioning is a branch of AI (Artificial Intelligence) technology that allows users to ask questions to an LLM (Large Language Model) using natural language. The LLM then analyzes and processes the data based on existing data and provides the results of the analysis.

[0003] Although LLMs can perform intelligent question processing based on databases and user-input questions or instructions, currently no LLM can achieve intelligent question processing based on user-input data table files such as Excel files. Summary of the Invention

[0004] In view of the above, embodiments of this application provide a table processing method, an electronic device, a storage medium, and a computer program product to at least partially solve the above problems.

[0005] According to a first aspect of the embodiments of this application, a table processing method is provided, comprising: processing a target table file through a generative model to obtain standard data table data corresponding to the target table file; storing the standard data table data in a target storage unit, and establishing an external table of the standard data table data in an analysis database corresponding to the target storage unit; processing the metadata of the external table through a generative model to generate a data query instruction; executing the data query instruction, and determining the data analysis result for the target table file based on the external table.

[0006] According to a second aspect of the embodiments of this application, a table processing method is provided, comprising: generating a table analysis request for a target table file based on an instruction input by a user from an instant messaging platform, wherein the instruction carries information about the target table file; sending the table analysis request to a server and receiving a data analysis result for the target table file generated by the server in response to the table analysis request, wherein the data analysis result is obtained by the server processing the target table file based on the obtained target table file through a generative model to obtain standard data table data corresponding to the target table file; storing the standard data table data in a target storage unit; establishing an external table of the standard data table data in an analysis database corresponding to the target storage unit; processing the metadata of the external table through a generative model to generate a data query instruction; executing the data query instruction; and determining a data analysis result based on the external table.

[0007] According to a third aspect of the embodiments of this application, an electronic device is provided, comprising: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store a computer program; and the processor is used to execute any of the methods described in the first and second aspects above by running the computer program stored in the memory.

[0008] According to a fourth aspect of the embodiments of this application, a computer storage medium is provided, on which a computer program is stored, which, when executed by a processor, implements the method described in either the first aspect or the second aspect.

[0009] According to a fifth aspect of the embodiments of this application, a computer program product is provided, including a computer program that, when executed by a processor, implements the method described in either the first aspect or the second aspect.

[0010] According to the table processing scheme provided in the embodiments of this application, the generative model can be used to effectively process the target table file and analyze the standard data table data corresponding to the target table file. The standard data table data can be stored in the target storage unit, and an external table of the standard data table data can be established in the analysis database corresponding to the target storage unit. Then, the generative model is used to process the metadata of the external table to generate data query instructions. By executing the data query instructions, a query is performed based on the external table to determine the data analysis results for the target table file. Thus, this scheme can effectively analyze the target table file by utilizing the generative model and external table query, effectively improve the intelligence of table file analysis, effectively realize the function of intelligent data query based on the user input of the target table file (such as an Excel file), and can better improve the user experience of viewing table files in various usage scenarios (such as including but not limited to instant messaging platforms). Attached Figure Description

[0011] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments recorded in the embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings.

[0012] Figure 1 This is a schematic diagram of an exemplary system to which the embodiments of this application are applicable.

[0013] Figure 2 This is a flowchart of the steps of a table processing method according to the first aspect of the embodiments of this application.

[0014] Figures 3A-3D This is a diagram illustrating some example table contents.

[0015] Figure 4 This is a flowchart of the steps of a table processing method according to a second aspect of an embodiment of this application.

[0016] Figure 5A A schematic diagram of a scenario according to an embodiment of this application is shown.

[0017] Figure 5B A schematic diagram illustrating the server-side operation process in one scenario according to an embodiment of this application is shown.

[0018] Figure 6 This is a schematic diagram of the structure of an electronic device according to an embodiment of this application. Detailed Implementation

[0019] To enable those skilled in the art to better understand the technical solutions in the embodiments of this application, the technical solutions in the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art should fall within the protection scope of the embodiments of this application.

[0020] Figure 1 An exemplary system applicable to embodiments of this application is shown. For example... Figure 1 As shown, the system 100 may include a cloud server 102, a communication network 104, and / or one or more user devices 106. Figure 1 The example in the text shows multiple user devices.

[0021] The cloud server 102 can be any suitable device for storing information, data, programs, and / or any other suitable type of content, including but not limited to distributed storage system devices, server clusters, computing cloud server clusters, etc. In some embodiments, the cloud server 102 can perform any suitable function. For example, in some embodiments, the cloud server 102 can determine the data analysis results for a target table file. For example, the cloud server 102 can process the target table file using a generative model to obtain standard data table data corresponding to the target table file; then store the standard data table data in the target storage unit, and establish an external table of the standard data table data in the analysis database corresponding to the target storage unit; then, based on the metadata of the external table, process it using a generative model to generate a data query instruction; then execute the data query instruction, and based on the external table, determine the data analysis results for the target table file. As an optional example, in some embodiments, a generative model is deployed in the cloud server 102. Optionally, after generating text, the cloud server 102 can return the generated data analysis results to the user device 106.

[0022] In some embodiments, the communication network 104 can be any suitable combination of one or more wired and / or wireless networks. For example, the communication network 104 can include any one or more of the following: the Internet, an intranet, a wide area network (WAN), a local area network (LAN), a wireless network, a digital subscriber line (DSL) network, a frame relay network, an asynchronous transfer mode (ATM) network, a virtual private network (VPN), and / or any other suitable communication network. The user equipment 106 can be connected to the communication network 104 via one or more communication links (e.g., communication link 112), and the communication network 104 can be linked to the cloud server 102 via one or more communication links (e.g., communication link 114). The communication link can be any communication link suitable for transmitting data between the user equipment 106 and the cloud server 102, such as a network link, a dial-up link, a wireless link, a hardwired link, any other suitable communication link, or any suitable combination of such links.

[0023] User device 106 may include any one or more user devices suitable for presenting text, interacting with users, etc. Optionally, user device 106 may be a client. As an optional example, user device 106 may be equipped with an instant messaging platform. In some embodiments, user device 106 may first generate a table analysis request for a target table file based on instructions input by the user from the instant messaging platform, wherein the instructions carry information about the target table file; then send the table analysis request to a server (which may be cloud server 102), and receive data analysis results for the target table file generated by the server in response to the table analysis request. In some embodiments, user device 106 may include any suitable type of device. For example, in some embodiments, user device 106 may include a mobile device, tablet computer, laptop computer, desktop computer, and / or any other suitable type of user device.

[0024] Based on the above system, this application provides a table processing solution, which will be described below through several embodiments.

[0025] Figure 2 This is a flowchart illustrating the steps of a table processing method according to a first aspect of an embodiment of this application. (Refer to...) Figure 2 As shown, the table processing method includes the following steps:

[0026] S202: Process the target table file using a generative model to obtain the standard data table data corresponding to the target table file.

[0027] It should be understood that the table processing method of the first aspect in this application embodiment can be used for table processing and analysis in any scenario. For example, it can be used for analyzing table files in an instant messaging platform. Optionally, the table analysis method of the first aspect can be executed by the server. For example, the server receives a table analysis request for a target table file generated based on the instructions entered by the user on the client's instant messaging platform, and then executes steps S202 to S204 in response to the table analysis request to obtain data analysis results, and returns the data analysis results to the client, thereby improving the user's experience of viewing the target table file through the client's instant messaging platform.

[0028] Optionally, the target table file in this application embodiment may be an Excel file (e.g., files with extensions such as .xlsx, .xlsm, .xls, etc.), or it may be other files such as Word documents containing tables.

[0029] Optionally, the generative model in this embodiment can be a Large Language Model (LLM). A Large Language Model (LLM) can be used to analyze tables.

[0030] In this embodiment, standard data table data can be data obtained by further processing standard table content data in the target table file. Correspondingly, non-standard data table data can be data obtained by further processing non-standard table content data in the target table file.

[0031] Because the valid data in the target spreadsheet file is presented in various and complex ways within its sheets, it can be divided into two categories according to certain rules: standard spreadsheet content data and non-standard spreadsheet content data. Standard spreadsheet content data refers to spreadsheet content data whose headers and rows can be accurately identified. Any other spreadsheet content data whose headers and rows cannot be accurately identified is considered non-standard spreadsheet content data. For example, standard spreadsheet content data can be referenced as follows... Figure 3A As illustrated in the example, the first row of this standard table content data is the table header, and the rows from the second row onwards are the table records. Table content data in this form, where the header and row records can be accurately identified, can be considered standard table content data.

[0032] For example, non-standard table content data can be referenced. Figures 3B-3DThe example shown illustrates that this non-standard table content data is diverse (including images, text, links, etc.), has an irregular layout, and contains merged cells. Similar table content data, including but not limited to this, in which the header and row records cannot be accurately determined, can be considered non-standard table content data.

[0033] Optionally, the target table file can be parsed first and converted to Markdown format for storage. In subsequent use, both standard and non-standard table content data can be in Markdown format. For example, after a user uploads a target table file (such as an Excel file), open-source tools like POI / Easy Excel can be used to parse the target table file and convert it to Markdown format for storage.

[0034] In some optional embodiments, step S102 may include: parsing the target table file page by page using a generative model to obtain the standard table content data of the target table file, and formatting the standard table content data to obtain the standard data table data corresponding to the target table file.

[0035] It should be understood that this approach leverages the capabilities of generative models to effectively parse the target table file page by page and then format it, thereby obtaining standard table data for easier subsequent processing.

[0036] For example, the target table file that has been converted to Markdown format can be input into a generative model (such as LLM). The generative model parses the Markdown target table file for each sheet to achieve page-by-page parsing, extracts the standard table content data, and then formats it to convert it into a certain required formatted data to obtain standard data table data for subsequent processing.

[0037] In this embodiment of the application, the formatting of standard table content data can be achieved through a generative model or by using other dedicated formatting programs, as long as the requirements are met.

[0038] In some optional embodiments, the standard table content data can be formatted in the following way to obtain the standard data table data corresponding to the target table file: extract the table header and each row of data from the standard table content data, store the extraction results as a CSV (Comma-Separated Values) format file, and determine the CSV format file as the standard data table data corresponding to the target table file.

[0039] For example, if the standard table content data of a single form (Sheet) can be parsed according to the standard data table format, then the table header and each row of data can be extracted row by row according to the standard data table format (the first row is the table header, and the rows from the second row onwards are the data table records). The results of each form extraction are stored in a CSV format file, and this CSV format file can be identified as the standard data table data corresponding to the target table file.

[0040] CSV stands for Comma-Separated Values, also known as Character-Separated Values ​​(because the separator character can be any character other than a comma). CSV files store tabular data (numbers and text) in plain text format. Plain text means the file is a sequence of characters and does not contain data that must be interpreted like binary numbers. A CSV file consists of any number of records, separated by some kind of newline character; each record consists of fields, separated by other characters or strings, most commonly commas or tabs. Typically, all rows have the exact same sequence of fields. CSV files are usually plain text files. Using CSV files to store and extract data as a standard tabular data format offers advantages such as ease of use, wide compatibility, human readability, high compression, lightweight nature, high flexibility, good cross-platform compatibility, and ease of data exchange and analysis.

[0041] For example, such as Figure 3A As shown, an example of storing it as a CSV file is as follows:

[0042] “

[0043] Record ID, order priority, discount, unit price, shipping cost, customer ID, customer name, shipping method, customer segment, and product category.

[0044] 1008,High,0,17,12,2189,Frank Cross,Regular Air,Corporate,Office Sup

[0045] 1041,Critical,0,301,24,3011,Tammy Ray,Regular Air,Corporate,OfficeSup

[0046] 1042,Critical,0,50,20,3011,Tammy Ray,Regular Air,Corporate,Technology

[0047] "

[0048] It should be understood that in this embodiment of the application, the table header and data of each row of the standard table content data are extracted line by line and stored as a CSV file. The CSV file is then stored as the standard data table data corresponding to the target table file. This allows the standard data table data to be implemented in plain text form and achieves advantages such as ease of use, wide compatibility, human readability, high compression, lightweight, high flexibility, good cross-platform compatibility, and ease of data exchange and analysis, thereby facilitating subsequent processing.

[0049] In other alternative embodiments, the target table file can be processed in ways other than generative models (such as LLM) to obtain standard and non-standard data table data corresponding to the target table file. For example, a judgment criterion can be pre-defined to first determine the standard and non-standard table content data of the target table file, and then format them separately to obtain the standard and non-standard data table data corresponding to the target table file.

[0050] S204: Store the standard data table data into the target storage unit, and create an external table of the standard data table data in the analysis database corresponding to the target storage unit.

[0051] In this embodiment, the target storage unit may be, but is not limited to, a database, cloud storage service, or memory. Step S204 establishes an external table of standard data in the corresponding analysis database within the target storage unit to facilitate subsequent data queries.

[0052] For example, in some optional embodiments, step S104 may include: storing standard data table data in the object storage service OSS, and creating an OSS external table of the standard data table data in the analytics database corresponding to the object storage service OSS.

[0053] Object Storage Service (OSS) is a massive, secure, low-cost, and highly reliable cloud storage service suitable for storing files of any type. It offers elastic scaling of capacity and processing power, multiple storage types to choose from, and comprehensive optimization of storage costs. It boasts high security, strong scalability, low complexity, ease of searching, and robust metadata management (each object can carry metadata, allowing users to easily set tags for objects, enabling flexible data management and retrieval), facilitating subsequent data queries.

[0054] Step S204 above can be implemented using any analytical database, such as an analytical database with OSS external linking capabilities. For example, in some optional embodiments, the analytical database can be AnalyticDB (ADB), which features high-concurrency read / write, low-peak-valley read / write, elastic scalability, and security and reliability. It can support petabyte-level data storage and is widely used in business intelligence, machine learning, real-time analytics, and data mining scenarios. ADB databases support rich SQL functionality, including UPDATE statements, which can be used for single-table updates and multi-table join updates.

[0055] Optionally, the OSS external table established in this embodiment can be an ADB data table created in an ADB database. ADB data tables are an important component of the cloud data warehouse ADB database. Through reasonable table type selection, table quantity management, and performance optimization measures, the high-performance data analysis service advantages of the ADB database can be fully utilized.

[0056] For example, in one feasible implementation, standard data table data in CSV file format can be stored in Object Storage Service (OSS) first, then an OSS external table corresponding to the standard data table data can be created in the ADB database, and the OSS external table can be linked with the standard data table data file stored in Object Storage Service (OSS).

[0057] It should be understood that the technical solutions in this application embodiment cleverly utilize the linking characteristics of the external database (such as ADB tables) of the Object Storage Service (OSS) for data storage and analysis, achieving efficient flow and processing of non-deterministic table structure data, while reducing storage costs and computing resource consumption. This facilitates data analysis of target table files and improves data analysis performance. Furthermore, the cloud architecture design of the Object Storage Service (OSS) and analysis database adopted in this application embodiment provides a solid foundation for large-scale data processing and represents an optimized integration of related technical frameworks, demonstrating promising application prospects.

[0058] S206: Metadata based on appearance is processed through a generative model to generate data query instructions.

[0059] Metadata, also known as intermediary data or relay data, is information about the organization of data, data domains, and their relationships. Simply put, metadata is data about data.

[0060] Optionally, data query instructions can be SQL (Structured Query Language) instructions. SQL is a special-purpose programming language, a database query and programming language used to access, query, update, and manage relational database systems. By generating data query instructions in the form of SQL commands, data queries can be effectively implemented, facilitating subsequent data analysis based on the data.

[0061] Optionally, in this embodiment of the application, metadata of the appearance (e.g., OSS appearance) can be obtained, prompt information corresponding to the metadata can be generated using Prompt Engineering, and then the prompt information can be analyzed and processed through a generative model to generate corresponding data query instructions.

[0062] For example, you can refer to the following example to understand the prompt message:

[0063]

[0064] The above prompts are analyzed and processed using a generative model, and the output may include data query commands (SQL commands).

[0065] S208: Execute a data query command to determine the data analysis results for the target table file based on the table form.

[0066] Optionally, the data query command (such as an SQL command) obtained in step S206 can be executed in the OSS external table (such as an ADB data table) to determine the data analysis results for the target table file. The data analysis results here can be the query results of the data query.

[0067] Based on this, in the embodiments of this application, through the optional implementation of steps S202 to S208 above, the generative model can be used to effectively process the target table file, analyze and obtain the standard data table data corresponding to the target table file, and store the standard data table data in the target storage unit. An external table of the standard data table data is established in the analysis database corresponding to the target storage unit. Then, the generative model processes the metadata of the external table to generate a data query instruction. By executing the data query instruction, a query is performed based on the external table to determine the data analysis results for the target table file. Thus, this solution can effectively realize the analysis of the target table file by utilizing the generative model and external table query, effectively improve the intelligence of table file analysis, effectively realize the function of intelligent data query based on the user input of the target table file (such as an Excel file), and can better improve the user experience of viewing table files in various usage scenarios (such as including but not limited to instant messaging platforms).

[0068] In some optional embodiments, step S108 may include: processing multiple different table files using a generative model to obtain standard data table data corresponding to each of the multiple table files, wherein the multiple different table files include a target table file. Obtaining standard data table data corresponding to multiple different table files facilitates subsequent multi-table joint queries. The specific implementation of step S208 is not specifically limited in this application embodiment. For example, in some optional embodiments, step S208 may include: executing a data query command, performing a joint query based on the external tables of the standard data table data corresponding to each of the multiple table files using a generative model, and determining the data analysis results for the target table file.

[0069] Optionally, the data query instructions (such as SQL instructions) obtained in step S206 can be executed in multiple different table files corresponding to OSS tables (such as ADB data tables), including the target table file, to achieve a joint query, so as to determine the data analysis results for the target table file.

[0070] It should be understood that the multi-table joint query method in this application embodiment can effectively improve the accuracy of data analysis results, thereby more effectively improving the intelligence of table file analysis. It can effectively realize the function of intelligent data query based on the user input of the target table file (such as Excel file), and can better improve the user experience of viewing table files in various usage scenarios (such as including but not limited to instant messaging platforms).

[0071] In some optional embodiments, the method of this application embodiment further includes: processing the data analysis results and metadata of the target table file through a generative model to generate an analysis report for the target table file.

[0072] Optionally, in this embodiment of the application, the data analysis results obtained in step S208 and the OSS table (such as an ADB data table) in step S206 are used as prompts in a generative model (such as an LLM) and input into the generative model for processing, so that the generative model generates an analysis report for the target table file.

[0073] For example, you can refer to the following example to understand the prompt message:

[0074]

[0075] By analyzing and processing the above prompts using a generative model, the output results can be obtained.

[0076] It should be understood that, in this embodiment of the application, by combining data analysis results and metadata of the table, and through the capabilities of generative models, an analysis report for the target table file can be effectively generated. This enables a more visual presentation of the data analysis results for the target table file, which is beneficial to improving the user experience and greatly improving the user's related work efficiency.

[0077] In addition, when this solution is applied to the table analysis scenario of instant messaging platforms, it allows users to obtain in-depth analysis of table files and corresponding analysis reports without leaving the chat interface. This immediacy can improve the user experience and greatly improve work efficiency, which is conducive to the intelligent upgrade and breakthrough of instant messaging platforms.

[0078] In this embodiment, generative models (such as LLM) can be effectively guided to generate customized data query instructions (such as SQL instructions) and analysis reports based on extracted standard data table data. This process is not only highly automated, but also allows for flexible adjustment of the analysis perspective according to data characteristics and user needs, effectively realizing advanced applications of artificial intelligence in the field of data analysis.

[0079] In some optional embodiments, after processing the target table file through a generative model in step S102, the table processing method of this application embodiment further includes: obtaining non-standard table content data corresponding to the target table file; formatting the non-standard table content data to obtain non-standard data table data of the target table file; and processing the data analysis results, metadata of the target table file, and non-standard data table data through a generative model to generate an analysis report for the target table file.

[0080] For example, such as Figures 3B-3D The example shown features non-standard table content data with diverse content (such as images, text, links, etc.), variable positions, and merged cells. Optionally, the target table file can be converted to Markdown format for storage (as mentioned earlier, open-source tools such as POI / Easy Excel can be used to parse the target table file and convert it to Markdown format for storage). The non-standard table content data corresponding to the obtained target table file can also be in Markdown format.

[0081] Non-standard tabular content data can be formatted in any way. For example, in some optional embodiments, non-standard tabular content data can be formatted using at least one of regular expression parsing, row-by-row parsing, and generative model parsing to obtain non-standard data table data in the target table file.

[0082] Among these, regular expression matching parsing refers to using preset regular expressions to parse non-standard tabular data to achieve formatting. Row-by-row reading parsing involves reading non-standard tabular data row by row and then parsing the results to achieve formatting. Generative model parsing refers to using a generative model (which could be LLM) to parse non-standard tabular data to achieve formatting.

[0083] You can choose any of the three methods mentioned above for formatting. For example, some implementations may use only one of these methods for formatting. Other implementations may use multiple methods simultaneously for formatting.

[0084] For example, some implementations may use multiple methods sequentially, continuing with the next method if the first method fails. For instance, some implementations may perform formatting in the order of regular expression matching, row-by-row parsing, and generative model parsing. For example, after obtaining non-standard table content data, regular expression matching can be performed first. If it succeeds, formatting is achieved, and row-by-row parsing and generative model parsing are no longer performed. If regular expression matching fails, row-by-row parsing is performed. If row-by-row parsing succeeds, formatting is achieved, and generative model parsing is no longer performed. If row-by-row parsing fails, generative model parsing is performed until parsing is successful, resulting in the non-standard table data of the target table file.

[0085] It should be understood that, in the embodiments of this application, at least one of the above three parsing methods can effectively format the standard table content data to obtain non-standard data table data of the target table file, so as to accurately generate an analysis report for the target table file in the future, and can improve the user's experience of viewing table files in various usage scenarios (such as including but not limited to instant messaging platforms).

[0086] After obtaining the non-standard data table data, in this embodiment of the application, the data analysis results obtained in step S208, the OSS table (such as the ADB data table) in step S206, and the obtained non-standard data table data can be used as prompts in a generative model (such as LLM) and input into the generative model for processing, so that the generative model can generate an analysis report for the target table file.

[0087] It should be understood that, on the one hand, in this embodiment, after processing the target table file through a generative model, non-standard table content data corresponding to the target table file can be obtained. This non-standard table content data is then formatted to obtain non-standard data table data for the target table file. This allows for adaptive data content recognition and acquisition, addressing the challenges of non-standard data formats in the target table file. Through collaboration between the generative model and intelligent algorithms, it can identify and process table content data with various complex formats (e.g., merged cells, irregular layouts, diverse content such as embedded media), achieving effective data extraction and formatting, thus facilitating better analysis of the target table file. On the other hand, by combining data analysis results, metadata of the table, and non-standard data table data, this embodiment, through the capabilities of the generative model, can effectively generate analysis reports for the target table file. This allows for a more visual presentation of the data analysis results for the target table file, improving user experience and significantly increasing user work efficiency.

[0088] Optionally, after receiving the analysis report for the target table file, the server can return the analysis report to the client, thereby meeting the user's need to view the analysis report of the target table file through the client's instant messaging platform and improving the user experience.

[0089] Optionally, the generative models used in the above steps in the embodiments of this application can be the same generative model or different generative models.

[0090] Figure 4 This is a flowchart illustrating the steps of a table processing method according to a second aspect of an embodiment of this application. (Refer to...) Figure 4 As shown, the table processing method includes the following steps:

[0091] S302: Based on the instructions input by the user from the instant messaging platform, generate a table analysis request for the target table file, wherein the instructions carry information about the target table file.

[0092] S304: Send a table analysis request to the server and receive the data analysis results for the target table file generated by the server in response to the table analysis request.

[0093] The data analysis results are generated by the server based on the acquired target table file, which is processed by a generative model to obtain the standard data table data corresponding to the target table file. The standard data table data is stored in the target storage unit, and an external table of the standard data table data is created in the analysis database corresponding to the target storage unit. Based on the metadata of the external table, the data is processed by a generative model to generate data query instructions, which are then executed. The data analysis results are determined based on the external table.

[0094] It should be understood that the table processing method of the second aspect of this application can be used on a client. The client can install an application of an instant messaging platform, which can be used to realize real-time table file processing and analysis. This solution can effectively realize the analysis of target table files by utilizing generative models and external queries, effectively improve the intelligence of table file analysis, effectively realize the function of intelligent querying based on user input of target table files (such as Excel files), and can significantly improve the user experience of viewing table files in the context of instant messaging platforms.

[0095] In some optional embodiments, the server performs page-by-page parsing on the target table file using a generative model to obtain the standard table content data of the target table file, and then formats the standard table content data to obtain the standard data table data corresponding to the target table file.

[0096] In some optional embodiments, the server extracts the header and each row of data from the standard table content data, stores the extraction results as a CSV file, and identifies the CSV file as the standard data table data corresponding to the target table file.

[0097] In some optional embodiments, the server stores standard table data in the Object Storage Service (OSS) and creates an OSS external table for the standard table data in the analytics database corresponding to the OSS.

[0098] In some optional embodiments, the server processes multiple different table files using a generative model to obtain standard data table data corresponding to each of the multiple table files, wherein the multiple different table files include a target table file; the server executes a data query command and performs a joint query based on the appearance of the standard data table data corresponding to each of the multiple table files using the generative model to determine the data analysis results for the target table file.

[0099] In some optional embodiments, the table processing method further includes: receiving an analysis report of the target table file generated by the server through a generative model based on the data analysis results and metadata of the target table file.

[0100] In some optional embodiments, the table processing method further includes: receiving an analysis report generated by the server based on the data analysis results of the target table file, the metadata of the table, and the non-standard data table data, and processing them through a generative model, wherein the non-standard data table data is obtained by the server from the non-standard table content data corresponding to the target table file and formatting the non-standard table content data.

[0101] In some optional embodiments, the server formats the non-standard table content data by at least one of regular expression parsing, row-by-row parsing, and generative model parsing to obtain the non-standard data table data of the target table file.

[0102] It should be understood that the relevant content and beneficial effects of the various optional embodiments of the table processing method of the second aspect of this application have been described in detail in the embodiments of the table processing method of the first aspect. For details, please refer to the relevant content above for understanding, and it will not be repeated here.

[0103] The following reference Figures 5A-5B The illustrated table processing solution provides an example scenario for understanding the table processing solution in this application embodiment. This scenario can be a table analysis scenario in an instant messaging platform.

[0104] like Figure 5A As shown, an instant messaging platform can be installed on the client. Within its chat page, any user can upload a target form file, such as... Figure 5A As shown, the target table file can be an Excel file with the .xlsx extension. The instant messaging platform's user interface can have corresponding prompts for the user to confirm whether to perform table analysis, such as... Figure 5A As shown, users can generate a command by triggering the "Quick Read: Intelligent Table Parsing" option. This command can carry information about the target table file. The client then generates a table analysis request for the target table file based on the command. The client can send the table analysis request to the server. After responding to the table analysis request, the server can obtain the data analysis results of the target table file and return them to the client. Users can then view the relevant data analysis results of the target table file on the client's instant messaging platform. It should be understood that... Figure 5A The analysis content shown in the table, "This table presents the XXX of different regional environments from different perspectives...", is only an example and is not intended to limit the embodiments of this application.

[0105] like Figure 5B As shown, this illustrates the server-side workflow for table processing in a given scenario. Figure 5BAs shown, the server can first convert the target table file uploaded by the user into Markdown format for storage. Then, it uses a generative model (which could be LLM) to parse it page by page, obtaining standard and non-standard table content data from the target table file. The standard content table data can be formatted to obtain standard data table data in CSV file format corresponding to the target table file. This CSV file-formatted standard data table data is then stored in Object Storage Service (OSS). An OSS external table for the standard data table data is then created in the corresponding ADB database (analysis database) of OSS, and linked to the standard data table data file stored in OSS. Afterwards, based on the metadata of the OSS external table, a generative model can be used to generate data query SQL commands. These SQL commands can then be executed based on the OSS external table to obtain data analysis results (for example, a single-table query can be performed on the OSS table corresponding to the target table file to obtain data analysis results; or a joint query can be performed on the OSS external tables corresponding to multiple different table files, including the target table file, to obtain data analysis results). Furthermore, at least one of the following methods—regular expression parsing, row-by-row parsing, and generative model parsing—can be used to format non-standard tabular content data, resulting in non-standard data table data (such as...) of the target table file. Figure 5B As shown, the non-standard table content data can first be parsed using regular expression matching. If parsing is successful, the non-standard data table data is obtained. If parsing fails, row-by-row parsing is performed. If parsing fails again, generative model parsing is used. Successful parsing yields the non-standard data table data. Then, based on the data analysis results of the target table file, its metadata, and the non-standard data table data, a generative model is used to process the data and generate an analysis report for the target table file. Both the data analysis results and the analysis report generated on the server can be returned to the client for viewing via the client's instant messaging platform.

[0106] Therefore, the technical solution in this application embodiment can effectively analyze the target table file by utilizing generative models and external queries, effectively improve the intelligence of table file analysis, effectively realize the function of intelligent querying based on user input of the target table file (such as Excel file), and can better improve the user experience in various usage scenarios (such as viewing table files in instant messaging platforms).

[0107] It should also be understood that the above Figure 5A and Figure 5BThe descriptions and related explanations are only for the purpose of facilitating understanding of the technical solutions of the embodiments of this application, and are not intended to limit the embodiments of this application in any way.

[0108] This application also provides an electronic device, including: a processor, a communication interface, a memory, and a communication bus, wherein the processor, the communication interface, and the memory communicate with each other through the communication bus; the memory is used to store a computer program; and the processor is used to execute any of the methods described in the foregoing multiple method embodiments by running the computer program stored in the memory.

[0109] Figure 6 A structural block diagram of an optional electronic device according to an embodiment of this application is shown. This application does not limit the specific implementation of the electronic device 1000; however, as an example, reference is made to... Figure 6 The electronic device 1000 provided in this application embodiment includes: a processor 1002, a communications interface 1004, a memory 1006, and a communication bus 1008. Wherein:

[0110] The processor 1002, communication interface 1004, and memory 1006 communicate with each other via communication bus 1008.

[0111] Communication interface 1004 is used to communicate with other electronic devices or servers.

[0112] The processor 1002 is used to execute the computer program 1010, specifically the relevant steps in any of the aforementioned method embodiments.

[0113] Specifically, computer program 1010 may include program code that includes computer operation instructions.

[0114] The processor 1002 may be a CPU, a GPU (Graphics Processing Unit), an Application Specific Integrated Circuit (ASIC), or one or more integrated circuits configured to implement the embodiments of this application. The one or more processors included in the smart device may be processors of the same type, such as one or more CPUs; or they may be processors of different types, such as one or more CPUs and one or more ASICs.

[0115] Memory 1006 is used to store computer program 1010. Memory 1006 may include high-speed RAM memory, and may also include non-volatile memory, such as at least one disk storage device.

[0116] Specifically, computer program 1010 can be used to cause processor 1002 to execute the method of any of the foregoing multiple method embodiments.

[0117] The specific implementation of each step in computer program 1010 can be found in the corresponding descriptions of the steps and units in any of the foregoing method embodiments, and will not be repeated here. Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the devices and modules described above can be referred to the corresponding process descriptions in the foregoing method embodiments, and will not be repeated here.

[0118] Furthermore, this application also provides a computer storage medium storing a computer program thereon, which, when executed by a processor, implements the methods of any of the foregoing method embodiments. The computer storage medium includes, but is not limited to, compact disc read-only memory (CD-ROM), random access memory (RAM), floppy disk, hard disk, or magneto-optical disk.

[0119] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the method as described in any of the above-described method embodiments.

[0120] Furthermore, it should be noted that the user-related information (including but not limited to user device information, user personal information, etc.) and data (including but not limited to sample data used for training the model, data used for analysis, stored data, displayed data, etc.) involved in the embodiments of this application are all information and data authorized by the user or fully authorized by all parties. Moreover, the collection, use and processing of related data must comply with the relevant laws, regulations and standards of the relevant countries and regions, and corresponding operation entry points are provided for users to choose to authorize or refuse.

[0121] It should be noted that, depending on the implementation needs, the various components / steps described in the embodiments of this application can be broken down into more components / steps, or two or more components / steps or parts of the operation of components / steps can be combined into new components / steps to achieve the purpose of the embodiments of this application.

[0122] The methods described in the embodiments of this application can be implemented in hardware, firmware, or as software or computer code that can be stored in a recording medium (such as a CD-ROM, RAM, floppy disk, hard disk, or magneto-optical disk), or as computer code downloaded over a network that is originally stored in a remote recording medium or a non-transitory machine-readable medium and will be stored in a local recording medium. Thus, the methods described herein can be stored on a recording medium using a general-purpose computer, a dedicated processor, or programmable or dedicated hardware (such as an Application Specific Integrated Circuit (ASIC) or a Field Programmable Gate Array (FPGA)). It is understood that the computer, processor, microprocessor controller, or programmable hardware includes storage components (e.g., Random Access Memory (RAM), Read-Only Memory (ROM), Flash Memory, etc.) capable of storing or receiving software or computer code, which, when accessed and executed by the computer, processor, or hardware, implements the methods described herein. Furthermore, when a general-purpose computer accesses code used to implement the methods shown herein, the execution of the code transforms the general-purpose computer into a dedicated computer for executing the methods shown herein.

[0123] Those skilled in the art will recognize that the units and method steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for specific applications, but such implementations should not be considered beyond the scope of the embodiments of this application.

[0124] The term "comprising" and its variations as used herein are open-ended, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". It should be noted that the concepts of "first", "second", etc., mentioned in the embodiments of this application are only used to distinguish different devices, modules, or units, and are not used to limit the order of functions performed by these devices, modules, or units or their interdependencies. It should be noted that the modifications of "a" and "a plurality" mentioned in the embodiments of this application are illustrative rather than restrictive, and those skilled in the art should understand that unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0125] The above embodiments are only used to illustrate the embodiments of this application, and are not intended to limit the embodiments of this application. Those skilled in the art can make various changes and modifications without departing from the spirit and scope of the embodiments of this application. Therefore, all equivalent technical solutions also fall within the scope of the embodiments of this application, and the patent protection scope of the embodiments of this application should be defined by the claims.

Claims

1. A table processing method, comprising: The target table file is processed by a generative model to obtain the standard data table data corresponding to the target table file; The standard data table data is stored in the target storage unit, and an external table of the standard data table data is created in the analysis database corresponding to the target storage unit. Based on the metadata of the appearance, data query instructions are generated through generative models. Execute the data query instruction and, based on the appearance, determine the data analysis results for the target table file.

2. The method according to claim 1, wherein, The process of processing the target table file using a generative model to obtain the standard data table data corresponding to the target table file includes: For the target table file, page-by-page parsing is performed using a generative model to obtain the standard table content data of the target table file. The standard table content data is then formatted to obtain the standard data table data corresponding to the target table file.

3. The method according to claim 2, wherein, The step of formatting the standard table content data to obtain the standard data table data corresponding to the target table file includes: For the standard table content data, extract the table header and each row of data row by row, store the extraction results as a CSV format file, and determine the CSV format file as the standard data table data corresponding to the target table file.

4. The method according to claim 1, wherein, The step of storing the standard data table data into the target storage unit and establishing an external table of the standard data table data in the analysis database corresponding to the target storage unit includes: The standard data table data is stored in the Object Storage Service (OSS), and an OSS external table of the standard data table data is created in the analysis database corresponding to the OSS.

5. The method according to claim 1, wherein, The step of processing the target table file through a generative model to obtain the standard data table data corresponding to the target table file includes: processing multiple different table files through a generative model to obtain the standard data table data corresponding to each of the multiple table files, wherein the multiple different table files include the target table file; The step of executing the data query command and determining the data analysis results for the target table file based on the appearance includes: executing the data query command, performing a joint query based on the appearance of the standard data table data corresponding to the multiple table files through a generative model, and determining the data analysis results for the target table file.

6. The method according to any one of claims 1-5, wherein, The method further includes: Based on the data analysis results of the target table file and the metadata of the table, a generative model is used to process the data and generate an analysis report for the target table file.

7. The method according to any one of claims 1-5, wherein, After processing the target table file using a generative model, the method further includes: Obtain the non-standard table content data corresponding to the target table file; The non-standard table content data is formatted to obtain the non-standard data table data of the target table file; Based on the data analysis results of the target table file, the metadata of the table, and the non-standard data table data, a generative model is used to process the data and generate an analysis report for the target table file.

8. The method according to claim 7, wherein, The step of formatting the non-standard table content data to obtain the non-standard data table data of the target table file includes: The non-standard table content data is formatted by at least one of regular expression parsing, row-by-row parsing, and generative model parsing to obtain the non-standard data table data of the target table file.

9. A table processing method, comprising: Based on the instructions input by the user from the instant messaging platform, a table analysis request for the target table file is generated, wherein the instructions carry information about the target table file; The system sends a table analysis request to the server and receives data analysis results for the target table file generated by the server in response to the table analysis request. The data analysis results are generated by the server processing the target table file using a generative model to obtain standard data table data corresponding to the target table file. The standard data table data is stored in a target storage unit. An external table of the standard data table data is created in the analysis database corresponding to the target storage unit. Based on the metadata of the external table, the system processes the metadata using a generative model to generate a data query instruction, executes the data query instruction, and determines the data analysis results based on the external table.

10. An electronic device, comprising: The processor, the communication interface, the memory, and the communication bus are provided, wherein the processor, the communication interface, and the memory communicate with each other via the communication bus. The memory is used to store computer programs; The processor is configured to perform the method of any one of claims 1-9 by running the computer program stored in the memory.

11. A computer storage medium having a computer program stored thereon, which, when executed by a processor, implements the method as described in any one of claims 1-9.

12. A computer program product comprising a computer program that, when executed by a processor, implements the method as described in any one of claims 1-9.

Images