Intelligent decision analysis method and system based on multi-source data fusion and dynamic template

Abstract

The invention relates to the field of data analysis, in particular to an intelligent decision analysis method and system based on multi-source data fusion and a dynamic template, and the method comprises the steps: receiving structured and unstructured data sources through an API gateway, executing data updating operation in a preset time period to obtain a data updating record, and storing the data updating record as an Excel file; converting the Excel file into a table structure according to sheet through an Excel-to-database engine, judging a field type in the table structure, and storing the field type in a database to obtain a field type database; constructing an analysis data set based on the field type database; generating a vector curve graph, a natural language query result and a time dimension word cloud according to the analysis data set; and carrying out confidence verification by utilizing a cross-engine verification chain, and binding the vector curve graph, the natural language query result and the time dimension word cloud by utilizing a dynamic association generator after verification is completed to obtain a decision analysis report. The decision analysis report generation efficiency and accuracy can be improved.

Description

Technical Field

[0001] The present invention belongs to the technical field of data analysis, and particularly relates to an intelligent decision-making analysis method and system based on multi-source data fusion and dynamic templates. Background Art

[0002] With the development of big data and artificial intelligence technologies, enterprises and institutions increasingly rely on data-driven analysis and reports in the decision-making process. However, existing decision support systems still have many limitations.

[0003] Currently, traditional decision-making systems usually rely on a single data source. For example, data acquisition and analysis are only based on a structured SQL database. In this mode, it is difficult for the system to interface with multi-source heterogeneous data (such as log data, sensor data, text data, image data, etc.). The data source compatibility is poor, and it is impossible to effectively integrate the access of hybrid data sources of domestic databases, open-source databases, APIs, and Excel, resulting in incomplete decision-making basis. At the level of analysis tools, existing technologies mostly use independent BI tools, statistical analysis tools, or scripting languages (such as the R language) for data processing and modeling. There is a lack of coordination mechanisms between these tools. Users often need to switch between multiple tools and cannot achieve cross-tool linked analysis, increasing the operation complexity. In terms of report generation, existing systems generally rely on manual integration. Users need to manually export charts, copy analysis results, and then paste them into a Word or Markdown template for typesetting. This method is not only time-consuming and laborious but also prone to omissions or errors, making it difficult to meet the requirements of efficient and accurate decision-making.

[0004] In addition, some studies have tried to introduce natural language processing technologies. For example, natural language queries are converted into database query statements through Text2SQL. However, such functions often lack connection with the report generation process and cannot achieve automatic embedding of analysis results and integrated report generation, resulting in low efficiency and accuracy in generating intelligent decision-making analysis reports. Summary of the Invention

[0005] The present invention provides an intelligent decision-making analysis method and system based on multi-source data fusion and dynamic templates, which can improve the efficiency and accuracy of generating decision-making analysis reports.

[0006] To achieve the above object, an intelligent decision-making analysis method based on multi-source data fusion and dynamic templates provided by the present invention includes: Receiving structured and unstructured data sources through an API gateway, configuring data source and interface information, and performing a data update operation within a preset time period to obtain a data update record, and storing the data update record as an Excel file; The Excel file is converted into a table structure by sheet using an Excel to database engine, and the field type database is obtained by using a deep learning model to determine the field type in the table structure and storing it in a preset database. Build and analyze datasets based on field type databases using the SQL dataset generator; Based on the analyzed dataset, the BI analysis engine generates embeddable vector graphics, and the predictive analysis engine uses a preset predictive model to predict trends and output vector curves. Based on the analyzed dataset, SQL statements are generated using a prompt word optimizer and Text2SQL in the conversational analysis engine, and the SQL statements are executed to obtain natural language query results. Based on the analysis dataset, word segmentation and word frequency statistics are performed in the hot word analysis engine to obtain word frequency statistics results, and a time-dimensional word cloud is generated based on the word frequency statistics results; The confidence level of vector curves, natural language query results, and time-dimensional word clouds is verified by using a cross-engine verification chain. After verification, a dynamic association generator is used to bind the vector curves, natural language query results, and time-dimensional word clouds to obtain a decision analysis report.

[0007] Optionally, the configuration data source and interface information includes: Configure the API gateway and assign a unique identifier to each data source, and associate the name, description and type of the data source with the corresponding unique identifier; Configure the interface request information, map the interface fields in the interface request information to the target table fields, and use a unique identifier to store the data source data into the preset data target table.

[0008] Optionally, the step of using a deep learning model to determine the field types in the table structure and storing them in a preset database to obtain a field type database includes: Create a database table based on the table structure name, and extract the field names from the database table row by row; The field type is obtained by using a deep learning model to predict the type of the field name. Bind field types to table structures and store the field types and table structures in a preset database to obtain a field type database.

[0009] Optionally, before obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the following steps are also included: Load the template containing variable placeholders, call the intelligent placeholder matching algorithm to parse the template structure, and obtain the template structure parsing result; Based on the template structure parsing results, a dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds. Then, a multimodal fusion engine is called to fill the template according to the placeholder type to obtain a decision analysis report.

[0010] Optionally, after obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the method further includes: embedding version traceability codes and invisible character watermarks into the decision analysis report to complete the protection of the decision analysis report.

[0011] To address the aforementioned problems, this invention also provides an intelligent decision analysis system based on multi-source data fusion and dynamic templates, the system comprising: The data preparation layer includes a heterogeneous data access module and an SQL dataset generator, wherein the heterogeneous data access module includes an API gateway and an Excel-to-database engine; The multimodal analysis layer includes a BI analysis engine, a predictive analysis engine, a conversational analysis engine, and a hot word analysis engine, and sets up a cross-engine verification chain; The dynamic report generation layer includes a dynamic correlation generator and a multimodal fusion engine.

[0012] Optionally, the heterogeneous data access module includes an API gateway and an Excel-to-database engine; The API gateway is used to configure the data source name and description, configure the interface request information of the data table, configure the interface field data structure and preview the data, and set the data update strategy. The Excel-to-database engine is used to create a new database in a domestic database, create corresponding data tables according to the sheet name, use the table header as a field, call a deep learning model to determine the field type and put the data into the database, thereby realizing the conversion of unstructured data into database tables.

[0013] Optionally, in the multimodal analysis layer: The BI analytics engine generates embeddable vector graphics; The predictive analytics engine uses a pre-defined predictive model to predict trends and outputs vector curves. The conversational analytics engine generates SQL statements based on instructions, data structures, user questions, reference examples, and other prompts, and executes the SQL statements to obtain natural language query results, which are output in the form of a conversational stream. The hot word analysis engine performs word segmentation and word frequency statistics to obtain word frequency statistics results, and generates a time-dimensional word cloud based on the word frequency statistics results.

[0014] Optionally, in the dynamic report generation layer, the dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds, and the multimodal fusion engine is used to generate decision analysis reports.

[0015] This invention utilizes an API gateway to receive structured and unstructured data sources, configures data source and interface information, and performs data update operations within a preset time period to obtain data update records. These records are then stored as Excel files, enabling unified and automatic updates of multi-source heterogeneous data. An Excel-to-database engine converts the Excel file into a table structure by sheet, and a deep learning model determines the field types within the table structure and stores them in a preset database, resulting in a field type database. This ensures the accuracy and consistency of the data structure. Furthermore, an SQL dataset generator constructs an analysis dataset based on the field type database, achieving centralized and standardized data resources. Additionally, by acquiring vector curves, natural language query results, and time-dimensional word clouds, the depth and dimensionality of data analysis are enhanced. Finally, a cross-engine validation chain is used to perform confidence verification on the vector curves, natural language query results, and time-dimensional word clouds. After verification, a dynamic association generator binds the vector curves, natural language query results, and time-dimensional word clouds to obtain a decision analysis report, improving the efficiency and accuracy of decision analysis report generation. Attached Figure Description

[0016] Figure 1 This is a flowchart illustrating an intelligent decision analysis method based on multi-source data fusion and dynamic templates provided in an embodiment of the present invention. Figure 2 This is a system module diagram of an intelligent decision analysis system based on multi-source data fusion and dynamic templates, provided as an embodiment of the present invention.

[0017] The realization of the objective, functional features and advantages of the present invention will be further explained in conjunction with the embodiments and with reference to the accompanying drawings. Detailed Implementation

[0018] It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.

[0019] This application provides an intelligent decision analysis method based on multi-source data fusion and dynamic templates. The executing entity of this intelligent decision analysis method based on multi-source data fusion and dynamic templates includes, but is not limited to, at least one of the following electronic devices that can be configured to execute the method provided in this application: a server, a terminal, or a computer. In other words, the intelligent decision analysis method based on multi-source data fusion and dynamic templates can be executed by software or hardware installed on a terminal device or a server device, and the software can be a blockchain platform. The server includes, but is not limited to, a single server, a server cluster, a cloud server, or a cloud server cluster. The server can be an independent server or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, network services, cloud communication, middleware services, domain name services, security services, content delivery networks (CDNs), and big data and artificial intelligence platforms.

[0020] Reference Figure 1 The diagram shown is a flowchart illustrating an intelligent decision analysis method based on multi-source data fusion and dynamic templates according to an embodiment of the present invention. In this embodiment, the intelligent decision analysis method based on multi-source data fusion and dynamic templates includes: S1. Utilize the API gateway to receive structured and unstructured data sources, configure data source and interface information, perform data update operations within a preset time period to obtain data update records, and store the data update records as an Excel file; Understandably, an API gateway is a unified service entry point used to receive data requests from external or internal systems. It can manage and schedule data interfaces from different sources in a unified manner, for example, periodically calling external API interfaces to obtain structured or unstructured data.

[0021] Understandably, structured data sources refer to data with fixed schemas, fields, and table structures, such as tables in relational databases used in government affairs (e.g., MySQL, Oracle, and the domestic database GaussDB).

[0022] Understandably, unstructured data refers to data without a fixed schema or table structure, such as Word documents of work plans, PDF documents of regulations, access log files of government websites that do not involve user privacy, on-site evidence photos of urban governance, meeting recordings, and promotional videos.

[0023] Understandably, interface information refers to the configuration information required when calling a data source API, such as the interface address (URL); request method (e.g., GET request, POST request); request parameters (e.g., time range, data type); authentication information (e.g., token, key, username, password); and returned data format (e.g., JSON, XML, CSV).

[0024] Understandably, the preset time period refers to the data update cycle set in advance; for example, the preset time period could be one month.

[0025] Specifically, the configuration data source and interface information includes: Configure the API gateway and assign a unique identifier to each data source, and associate the name, description and type of the data source with the corresponding unique identifier; Configure the interface request information, map the interface fields in the interface request information to the target table fields, and use a unique identifier to store the data source data into the preset data target table.

[0026] Understandably, a unique identifier is a code or mark that the system assigns to each data source, which is unique globally and is used to uniquely identify a particular data source.

[0027] S2. The Excel file is converted into a table structure by sheet using an Excel to database engine, and the field types in the table structure are determined by a deep learning model and stored in a preset database to obtain a field type database.

[0028] Understandably, an Excel-to-database engine refers to a tool that automatically parses the contents of an Excel file and converts them into database tables.

[0029] Understandably, a sheet refers to a worksheet in an Excel file, and each sheet can be considered an independent two-dimensional table.

[0030] Understandably, table structure refers to the definition of a table in a database, including metadata such as table name, field names, field types, and constraints.

[0031] Understandably, a deep learning model refers to an algorithm based on neural networks; for example, a deep learning model could be the BERT algorithm model.

[0032] Specifically, the step of using a deep learning model to determine the field types in the table structure and storing them in a preset database to obtain a field type database includes: Create a database table based on the table structure name, and extract the field names from the database table row by row; The field type is obtained by using a deep learning model to predict the type of the field name. Bind field types to table structures and store the field types and table structures in a preset database to obtain a field type database.

[0033] S3. Construct an analysis dataset based on the field type database using the SQL dataset generator.

[0034] Understandably, an SQL dataset generator is a tool used to automatically generate, combine, or extract datasets based on existing field type databases, and construct datasets in the form of SQL statements that can be directly used for analysis, modeling, or testing.

[0035] S4. Based on the analysis dataset, use the BI analysis engine to generate embeddable vector graphics, and use the preset prediction model in the predictive analysis engine to perform trend prediction and output vector curves.

[0036] Understandably, a BI analytics engine refers to a core component of Business Intelligence (BI) that can process, model, and visualize analytical datasets to generate interactive charts (such as vector graphs).

[0037] S5. Based on the analyzed dataset, use the prompt word optimizer and Text2SQL in the conversational analysis engine to generate SQL statements, and execute the SQL statements to obtain natural language query results.

[0038] Understandably, a conversational analytics engine is an analytics module that combines natural language processing and database query technology. Users ask questions through natural language dialogue, and the system uses prompt word optimizer and Text2SQL technology to convert the questions into SQL statements, execute them, and then return the results in natural language.

[0039] S6. Based on the analysis dataset, perform word segmentation and word frequency statistics in the hot word analysis engine to obtain word frequency statistics results, and generate a time-dimensional word cloud based on the word frequency statistics results.

[0040] Understandably, a hot word analysis engine refers to a text analysis module used to segment and count word frequencies in an analysis dataset, identify and extract keywords, and identify high-frequency words.

[0041] Understandably, time-dimensional word clouds refer to ordinary word clouds that incorporate time series to visualize the changes in word frequency over different time periods, showing the evolution of keyword popularity over time.

[0042] S7. Utilize cross-engine verification chains to perform confidence verification on vector curves, natural language query results, and time-dimensional word clouds. After verification, use a dynamic association generator to bind vector curves, natural language query results, and time-dimensional word clouds to obtain a decision analysis report.

[0043] Understandably, cross-engine verification chains refer to a mechanism that cross-verifies the consistency and reliability of results through multiple different analytics engines.

[0044] Understandably, confidence level is a quantitative indicator of the degree to which an analysis result or prediction result is credible, usually expressed as a probability or percentage.

[0045] Understandably, a dynamic association generator is an automated module used to dynamically bind analytical results from different sources or in different forms (such as charts, text, and word clouds) semantically or logically to generate a unified decision analysis report.

[0046] Specifically, before obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the following steps are also included: Load the template containing variable placeholders, call the intelligent placeholder matching algorithm to parse the template structure, and obtain the template structure parsing result; Based on the template structure parsing results, a dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds. Then, a multimodal fusion engine is called to fill the template according to the placeholder type to obtain a decision analysis report.

[0047] Understandably, the placeholder intelligent matching algorithm refers to a template parsing algorithm that automatically identifies the position, type, and meaning of variable placeholders in a template, matches them with database fields or external variables, and generates an executable template structure parsing result.

[0048] Furthermore, after obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the method further includes: embedding version traceability codes and invisible character watermarks into the decision analysis report to complete the protection of the decision analysis report.

[0049] Understandably, version traceability coding refers to a unique identifier embedded in a document or report to mark information such as the document's version source, generation time, and responsible party, thereby enabling version identification and traceability when the document is disseminated, modified, or leaked.

[0050] Understandably, invisible character watermarking refers to a digital watermarking technology that uses zero-width characters, unprintable characters, or special encoding to conceal identification information within document text. It is invisible to the naked eye and does not affect normal reading, but can be extracted through specialized algorithms for copyright protection, traceability, and anti-tampering purposes.

[0051] This invention utilizes an API gateway to receive structured and unstructured data sources, configures data source and interface information, and performs data update operations within a preset time period to obtain data update records. These records are then stored as Excel files, enabling unified and automatic updates of multi-source heterogeneous data. An Excel-to-database engine converts the Excel file into a table structure by sheet, and a deep learning model determines the field types within the table structure and stores them in a preset database, resulting in a field type database. This ensures the accuracy and consistency of the data structure. Furthermore, an SQL dataset generator constructs an analysis dataset based on the field type database, achieving centralized and standardized data resources. Additionally, by acquiring vector curves, natural language query results, and time-dimensional word clouds, the depth and dimensionality of data analysis are enhanced. Finally, a cross-engine validation chain is used to perform confidence verification on the vector curves, natural language query results, and time-dimensional word clouds. After verification, a dynamic association generator binds the vector curves, natural language query results, and time-dimensional word clouds to obtain a decision analysis report, improving the efficiency and accuracy of decision analysis report generation.

[0052] like Figure 2 The diagram shown is a system module diagram of an intelligent decision analysis system based on multi-source data fusion and dynamic templates provided in an embodiment of the present invention.

[0053] Specifically, the system includes: The data preparation layer includes a heterogeneous data access module and an SQL dataset generator, wherein the heterogeneous data access module includes an API gateway and an Excel-to-database engine; The multimodal analysis layer includes a BI analysis engine, a predictive analysis engine, a conversational analysis engine, and a hot word analysis engine, and sets up a cross-engine verification chain; The dynamic report generation layer includes a dynamic correlation generator and a multimodal fusion engine.

[0054] Furthermore, the heterogeneous data access module includes an API gateway and an Excel-to-database engine; The API gateway is used to configure the data source name and description, configure the interface request information of the data table, configure the interface field data structure and preview the data, and set the data update strategy. The Excel-to-database engine is used to create a new database in a domestic database, create corresponding data tables according to the sheet name, use the table header as a field, call a deep learning model to determine the field type and put the data into the database, thereby realizing the conversion of unstructured data into database tables.

[0055] Furthermore, in the multimodal analysis layer: The BI analytics engine generates embeddable vector graphics; The predictive analytics engine uses a pre-defined predictive model to predict trends and outputs vector curves. The conversational analytics engine generates SQL statements based on instructions, data structures, user questions, reference examples, and other prompts, and executes the SQL statements to obtain natural language query results, which are output in the form of a conversational stream. The hot word analysis engine performs word segmentation and word frequency statistics to obtain word frequency statistics results, and generates a time-dimensional word cloud based on the word frequency statistics results.

[0056] Furthermore, in the dynamic report generation layer, the dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds, and the multimodal fusion engine is used to generate decision analysis reports.

[0057] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs.

[0058] Furthermore, the functional modules in the various embodiments of the present invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated unit can be implemented in hardware or in the form of hardware plus software functional modules.

[0059] It will be apparent to those skilled in the art that the present invention is not limited to the details of the exemplary embodiments described above, and that the present invention can be implemented in other specific forms without departing from the spirit or essential characteristics of the present invention.

[0060] Therefore, the embodiments should be considered exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of equivalents of the claims are intended to be embraced within the invention. No appended diagram markings in the claims should be construed as limiting the scope of the claims.

[0061] The blockchain referred to in this invention is a novel application model of computer technologies such as distributed data storage, peer-to-peer transmission, consensus mechanisms, and encryption algorithms. Essentially, a blockchain is a decentralized database, a chain of data blocks linked together using cryptographic methods. Each data block contains information about a batch of network transactions, used to verify the validity of the information (anti-counterfeiting) and generate the next block. A blockchain can include an underlying blockchain platform, a platform product service layer, and an application service layer.

[0062] The embodiments of this application can acquire and process relevant data based on artificial intelligence technology. Artificial intelligence (AI) refers to the theories, methods, technologies, and application systems that use digital computers or machines controlled by digital computers to simulate, extend, and expand human intelligence, perceive the environment, acquire knowledge, and use that knowledge to obtain optimal results.

[0063] Furthermore, it is clear that the word "comprising" does not exclude other units or steps, and the singular does not exclude the plural. Multiple units or devices recited in a system claim may also be implemented by a single unit or device through software or hardware. The terms "first," "second," etc., are used to indicate names and do not indicate any specific order.

[0064] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention.

Claims

1. An intelligent decision analysis method based on multi-source data fusion and dynamic templates, characterized in that, The method includes: The API gateway is used to receive structured and unstructured data sources, configure data source and interface information, perform data update operations within a preset time period to obtain data update records, and store the data update records as an Excel file. The Excel file is converted into a table structure by sheet using an Excel to database engine, and the field type database is obtained by using a deep learning model to determine the field type in the table structure and storing it in a preset database. Build and analyze datasets based on field type databases using the SQL dataset generator; Based on the analyzed dataset, the BI analysis engine generates embeddable vector graphics, and the predictive analysis engine uses a preset predictive model to predict trends and output vector curves. Based on the analyzed dataset, SQL statements are generated using a prompt word optimizer and Text2SQL in the conversational analysis engine, and the SQL statements are executed to obtain natural language query results. Based on the analysis dataset, word segmentation and word frequency statistics are performed in the hot word analysis engine to obtain word frequency statistics results, and a time-dimensional word cloud is generated based on the word frequency statistics results; The confidence level of vector curves, natural language query results, and time-dimensional word clouds is verified by using a cross-engine verification chain. After verification, a dynamic association generator is used to bind the vector curves, natural language query results, and time-dimensional word clouds to obtain a decision analysis report.

2. The intelligent decision analysis method based on multi-source data fusion and dynamic templates as described in claim 1, characterized in that, The configuration data source and interface information includes: Configure the API gateway and assign a unique identifier to each data source, and associate the name, description and type of the data source with the corresponding unique identifier; Configure the interface request information, map the interface fields in the interface request information to the target table fields, and use a unique identifier to store the data source data into the preset data target table.

3. The intelligent decision analysis method based on multi-source data fusion and dynamic templates as described in claim 1, characterized in that, The process of using a deep learning model to determine the field types in a table structure and storing them in a preset database, resulting in a field type database, includes: Create a database table based on the table structure name, and extract the field names from the database table row by row; The field type is obtained by using a deep learning model to predict the type of the field name. Bind field types to table structures and store the field types and table structures in a preset database to obtain a field type database.

4. The intelligent decision analysis method based on multi-source data fusion and dynamic templates as described in claim 1, characterized in that, Before obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the process also includes... Load the template containing variable placeholders, call the intelligent placeholder matching algorithm to parse the template structure, and obtain the template structure parsing result; Based on the template structure parsing results, a dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds. Then, a multimodal fusion engine is called to fill the template according to the placeholder type to obtain a decision analysis report.

5. The intelligent decision analysis method based on multi-source data fusion and dynamic templates as described in claim 1 or 4, characterized in that, After obtaining the decision analysis report by binding vector curves, natural language query results, and time-dimensional word clouds using a dynamic association generator, the process further includes: embedding version traceability codes and invisible character watermarks into the decision analysis report to complete the protection of the decision analysis report.

6. An intelligent decision analysis system based on multi-source data fusion and dynamic templates, characterized in that, The system is used to implement the intelligent decision analysis method based on multi-source data fusion and dynamic templates as described in any one of claims 1 to 7, and the system includes: The data preparation layer includes a heterogeneous data access module and an SQL dataset generator, wherein the heterogeneous data access module includes an API gateway and an Excel-to-database engine; The multimodal analysis layer includes a BI analysis engine, a predictive analysis engine, a conversational analysis engine, and a hot word analysis engine, and sets up a cross-engine verification chain; The dynamic report generation layer includes a dynamic correlation generator and a multimodal fusion engine.

7. The intelligent decision analysis system based on multi-source data fusion and dynamic templates as described in claim 6, characterized in that, The heterogeneous data access module includes an API gateway and an Excel-to-database engine; The API gateway is used to configure the data source name and description, configure the interface request information of the data table, configure the interface field data structure and preview the data, and set the data update strategy. The Excel-to-database engine is used to create a new database in a domestic database, create corresponding data tables according to the sheet name, use the table header as a field, call a deep learning model to determine the field type and put the data into the database, thereby realizing the conversion of unstructured data into database tables.

8. The intelligent decision analysis system based on multi-source data fusion and dynamic templates as described in claim 6, characterized in that, In the multimodal analysis layer: The BI analytics engine generates embeddable vector graphics; The predictive analytics engine uses a pre-defined predictive model to predict trends and outputs vector curves. The conversational analytics engine generates SQL statements based on instructions, data structures, user questions, reference examples, and other prompts, and executes the SQL statements to obtain natural language query results, which are output in the form of a conversational stream. The hot word analysis engine performs word segmentation and word frequency statistics to obtain word frequency statistics results, and generates a time-dimensional word cloud based on the word frequency statistics results.

9. The intelligent decision analysis system based on multi-source data fusion and dynamic templates as described in claim 6, characterized in that, In the dynamic report generation layer, the dynamic association generator is used to bind vector curves, natural language query results, and time-dimensional word clouds, and the multimodal fusion engine is used to generate decision analysis reports.

Images