Automatic benchmarking method and system based on large model and storable medium

Abstract

The invention provides an automatic benchmarking method and system based on a large model and a storable medium, in response to a table name input by a user, the system transmits metadata and sample data of a resource table corresponding to the table name to the large model; the large model generates semantic interpretation corresponding to each field in the resource table to obtain a recognition result, and the system stores the recognition result into a database original table corresponding to the resource table as a newly added column and splits the resource table into field arrays according to the fields; traversing the field array by a system, and comparing elements in the array with data elements in a vector knowledge base by utilizing an RAG technology to obtain data elements similar to the field array; the large model selects similar data elements for benchmarking according to the metadata to obtain a benchmarking result of the current resource table, and meanwhile the benchmarking result serves as a newly-added column to be stored in the database original table. According to the method, a manually maintained hard coding rule is converted into a semantic matching model, the method has the advantages of self-adaptability, intelligence and lightweight operation and maintenance, and the data value can be released.

Description

Technical Field

[0001] This application relates to the field of data processing technology, specifically to an automated benchmarking method, system, and storage medium based on a large model. Background Technology

[0002] With the growing demand for data sharing and data mining, big data governance plays a crucial role across various industries. Data elements are a key means of transforming diverse and heterogeneous data into standardized data, and data element benchmarking is a critical step in data governance. As the application scope of data governance expands, different industries have different data standards, and the same field can represent different semantics in different contexts. Data governance relies heavily on the deep involvement of domain experts, making the process time-consuming and difficult to scale. Therefore, the ability to quickly and effectively conduct cross-industry data benchmarking is particularly important.

[0003] Most current automated benchmarking methods define data mapping rules and thresholds through configuration files such as YAML / INI, requiring metadata matching to achieve comparison. However, current automated benchmarking methods have certain limitations:

[0004] First, the existing methods lack flexibility and adaptability. They are difficult to dynamically adjust rule thresholds according to the scenario, and the stability of automated scripts is poor when faced with resource tables that fail to load.

[0005] Secondly, data integration and analysis are insufficient. Existing benchmarking methods often require manual maintenance of mapping relationships, making it difficult to automatically form mapping relationships based on contextual semantics, which often limits benchmarking efficiency.

[0006] Furthermore, the maintenance cost of data benchmarking is high, and changes in the data benchmarking mapping relationship require frequent script updates, which will increase the burden of operation and maintenance.

[0007] To address the above issues, an automated benchmarking method, system, and storage medium based on a large model are proposed. Summary of the Invention

[0008] This application addresses one or more technical deficiencies in the prior art by proposing the following technical solution.

[0009] Based on the first aspect of this application, an automated benchmarking method based on large models is proposed, including:

[0010] S1: In response to the table name entered by the user, the system passes the metadata and sample data of the resource table corresponding to the table name to the large model;

[0011] S2: The large model generates semantic interpretations corresponding to each field in the resource table to obtain recognition results. The system stores the recognition results as new columns in the original database table corresponding to the resource table and splits the resource table into field arrays according to the fields.

[0012] S3: The system iterates through the field array and uses RAG technology to compare the elements in the field array with the data elements in the vector knowledge base to obtain similar data elements in the field array; and

[0013] S4: The large model selects similar data elements for benchmarking based on the metadata, obtains the benchmarking result of the current resource table and outputs it, and at the same time stores the benchmarking result as a new column in the original database table.

[0014] This application provides an end-to-end automated benchmarking method that replaces traditional manual rule configuration with the semantic understanding capabilities of large models, greatly improving the intelligence and automation level of the benchmarking process, significantly reducing manual intervention, and improving the processing efficiency and accuracy of the benchmarking process.

[0015] Furthermore, prior to S1, standardized data elements and metadata specific to the industry are stored in a vector knowledge base, with each data element treated as an independent vectorized segment.

[0016] Building a vector knowledge base enables modular and standardized processing of benchmarking knowledge, allowing for rapid and flexible adaptation to different industry standards.

[0017] Furthermore, after the system passes the metadata and sample data of the corresponding resource table to the large model, if the parameters passed in the resource table are empty, then the resource table does not exist and an exception is thrown.

[0018] The parameter validation and exception handling mechanism can promptly identify and report invalid inputs, preventing the process from terminating due to dirty or erroneous data input, thus ensuring the stability and reliability of the system operation.

[0019] Furthermore, the RAG technology is used to compare the elements in the array with the data elements in the vector knowledge base, including: performing a similarity search in the vector knowledge base based on the field name, data type, and the recognition result, and recalling one or more candidate data elements.

[0020] This step optimizes the accuracy of the retrieval, enabling a more comprehensive and accurate understanding of the field intent and recalling more relevant and higher-quality candidate data elements.

[0021] Furthermore, it also includes inputting field information and its context, along with the recalled candidate data elements, into a large model, whereby the large model selects the most matching data element as the benchmark result for that field.

[0022] This step fully leverages the contextual understanding and reasoning capabilities of large models, intelligently selecting the globally optimal solution from multiple similar options. This overcomes the bias inherent in simple vector similarity matching, greatly improving the accuracy and reliability of the final benchmarking results.

[0023] Furthermore, the recognition results are automatically generated by the large model based on field names, sample data, and contextual semantics, and are used to expand the semantic representation of the fields, including synonyms, near-synonyms, or explanatory terms for each field.

[0024] This step fundamentally solves the matching failure problem caused by name ambiguity, abbreviations, industry jargon, etc., and is one of the key innovations of this method to achieve high-precision semantic matching.

[0025] Based on a second aspect of this application, a system for benchmarking using the method described in any of the above claims is also proposed, comprising a timed task scheduler and a resource table automatic benchmarking module;

[0026] The scheduled task scheduler is configured to perform timed matching of the resource tables of the automatic resource table matching module.

[0027] The resource table automatic alignment module includes a metadata intelligent parsing module, a vector knowledge base, and a large model-assisted alignment module.

[0028] The vector knowledge base is used to store standardized data elements and metadata for a specific industry. The metadata intelligent parsing module parses the metadata information in the vector knowledge base to obtain the parsing results. The large model-assisted benchmarking module uses the parsing results and, in conjunction with RAG technology, performs automated benchmarking from the original database tables to standardized data elements.

[0029] Based on a third aspect of this application, a computer program product is also proposed, having one or more computer programs thereon that, when executed by a computer processor, perform the method as described in any of the above.

[0030] The technical advantages of this application are as follows: This application proposes an automatic benchmarking method, system, and storage medium based on a large model. By storing industry data element standards through a vector knowledge base and combining the semantic understanding capabilities of the large model, it achieves dynamic threshold adjustment, intelligent mapping generation, and rapid cross-industry adaptation. Compared with the traditional data benchmarking method based on configuration files, its core value lies in transforming manually maintained hard-coded rules into a learnable semantic matching model, enabling the system to have the advantages of adaptability, intelligence, and lightweight operation and maintenance. It is suitable for scenarios with inconsistent standards and complex heterogeneous systems in digital transformation, and can further assist in data governance and release data value. Attached Figure Description

[0031] Other features, objects, and advantages of this application will become more apparent from the following detailed description of non-limiting embodiments with reference to the accompanying drawings.

[0032] Figure 1 This is a flowchart illustrating an automated benchmarking method based on a large model, according to an embodiment of this application.

[0033] Figure 2 This is a framework diagram of an automated benchmarking system based on a large model, provided according to an embodiment of this application.

[0034] Figure 3 This is a flowchart of a resource table benchmarking process for an automated benchmarking system based on a large model, provided according to an embodiment of this application.

[0035] Figure 4 This is an example diagram of data element segmentation for an automated benchmarking system based on a large model, provided according to an embodiment of this application.

[0036] Figure 5 This is an example diagram illustrating the conversion of resource table fields to arrays in an automated benchmarking system based on a large model, according to an embodiment of this application.

[0037] Figure 6 This is an example diagram of data item benchmarking for an automated benchmarking system based on a large model, provided according to an embodiment of this application.

[0038] Figure 7 This is an example diagram illustrating the collation of collation results in an automated collation system based on a large model, according to an embodiment of this application.

[0039] Figure 8 This is a schematic diagram of the structure of a computer system suitable for implementing the electronic devices of the present application embodiments. Detailed Implementation

[0040] The present application will now be described in further detail with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and not intended to limit it. Furthermore, it should be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings.

[0041] It should be noted that, unless otherwise specified, the embodiments and features described in this application can be combined with each other. This application will now be described in detail with reference to the accompanying drawings and embodiments.

[0042] Figure 1 An automated benchmarking method based on a large model is shown, including:

[0043] S1: In response to the table name entered by the user, the system passes the metadata and sample data of the resource table corresponding to the table name to the large model;

[0044] S2: The large model generates semantic interpretations corresponding to each field in the resource table to obtain recognition results. The system stores the recognition results as new columns in the original database table corresponding to the resource table and splits the resource table into field arrays according to the fields.

[0045] S3: The system iterates through the field array and uses RAG technology to compare the elements in the field array with the data elements in the vector knowledge base to obtain similar data elements in the field array; and

[0046] S4: The large model selects similar data elements for benchmarking based on the metadata, obtains the benchmarking result of the current resource table and outputs it, and at the same time stores the benchmarking result as a new column in the original database table.

[0047] It should be noted that this application provides an end-to-end automated benchmarking method, which replaces traditional manual rule configuration with the semantic understanding capabilities of large models, greatly improving the intelligence and automation level of the benchmarking process, reducing manual intervention, and improving the processing efficiency and accuracy of the benchmarking process.

[0048] It should be noted that before S1, standardized data elements and metadata for specific industries are stored in a vector knowledge base, and each data element is treated as an independent vectorized segment.

[0049] It should be noted that building a vector knowledge base enables modular and standardized processing of benchmark knowledge, allowing for rapid and flexible adaptation to different industry standards.

[0050] It should be noted that after the system passes the metadata and sample data of the corresponding resource table to the large model, if the parameters passed in the resource table are empty, the resource table does not exist and an exception is thrown.

[0051] It should be noted that the parameter validation and exception handling mechanism can promptly identify and report invalid inputs, preventing the process from terminating due to dirty or erroneous data input, thus ensuring the stability and reliability of the system operation.

[0052] It should be noted that the RAG technology is used to compare the elements in the array with the data elements in the vector knowledge base, including: based on the field name, data type and the recognition result, performing a similarity search in the vector knowledge base to recall one or more candidate data elements.

[0053] It should be noted that this step optimizes the accuracy of the retrieval, enabling a more comprehensive and accurate understanding of the field intent and recalling more relevant and higher-quality candidate data elements.

[0054] It should be noted that this also includes inputting field information and its context, along with the recalled candidate data elements, into the large model, whereby the large model selects the most matching data element as the benchmark result for that field.

[0055] It should be noted that this step fully utilizes the contextual understanding and reasoning capabilities of the large model, which can intelligently filter out the globally optimal solution from multiple similar options, overcoming the biases present in simple vector similarity matching, and greatly improving the accuracy and reliability of the final benchmarking results.

[0056] It should be noted that the recognition results are automatically generated by the large model based on field names, sample data, and contextual semantics, and are used to expand the semantic representation of the fields, including synonyms, near-synonyms, or explanatory terms for each field.

[0057] It should be noted that this step fundamentally solves the matching failure problem caused by name ambiguity, abbreviations, industry jargon, etc., and is one of the key innovations of this method to achieve high-precision semantic matching.

[0058] The following is for reference. Figure 2 , Figure 2 A framework diagram of an automated benchmarking system based on a large model is shown, including a timed task scheduler and an automatic resource table benchmarking module.

[0059] In a specific embodiment, the scheduled task scheduler is configured to perform timed matching of the resource table of the automatic resource table matching module.

[0060] In a specific embodiment, the resource table automatic benchmarking module includes a metadata intelligent parsing module, a vector knowledge base, and a large model-assisted benchmarking module;

[0061] The vector knowledge base is used to store standardized data elements and metadata for a specific industry. The metadata intelligent parsing module parses the metadata information in the vector knowledge base to obtain the parsing results. The large model-assisted benchmarking module uses the parsing results and, in conjunction with RAG technology, performs automated benchmarking from the original database tables to standardized data elements.

[0062] In a specific embodiment, the automatic matching process of the resource table is as follows: Figure 3 As shown in Table 1, common data elements for bank information systems are set up and stored in the vector knowledge base. The selection is to segment by row, dividing the relevant content of a data element into a segment, such as... Figure 4 As shown;

[0063] Table 1 General Data Elements for Bank Information Technology

[0064]

[0065]

[0066] The user inputs the resource table name, retrieves the resource to be analyzed from the system database. The metadata of the resource to be analyzed is shown in Table 2, and the sample data is shown in Table 3. The contents of Table 2 and Table 3 are passed to the big data model, which generates the semantic interpretation of each field in the resource table and obtains the recognition result, as shown in the last column of Table 4.

[0067] Table 2 Bank User Account Opening Information - Metadata Table

[0068] Data Item ID Chinese name of data item English name Data item type Remark 11320030000001-001 Customer Number Customernumber string — 11320030000001-002 Name Fullname string — 11320030000001-003 Name in pinyin Pinyinofname string — 11320030000001-004 gender Gender string — 11320030000001-004 Document Types Typeofdocument string — 11320030000001-005 ID number Documentnumber string — 11320030000001-006 Contact number Contact number string — 11320030000001-007 Customer types Customertype string — 11320030000001-008 User Status Userstatus string — 11320030000001-009 account Accountnumber string — 11320030000001-010 Account type Accounttype string — 11320030000001-011 Contact address Address string — 11320030000001-012 Account Manager Name Name of customermanager string — 11320030000001-013 Account Manager Number Customermanager number string — 11320030000001-014 Account Manager Contact Number Customermanager contact number string — 11320030000001-015 Branch of Household Management Branch string — 11320030000001-016 Branch where the account was opened Openingbranch string — 11320030000001-017 Account Manager Name Openingcustomermanagername string — 11320030000001-018 Account opening manager number Opening customermanagernumber string — 11320030000001-019 Account opening time Opening time date — 11320030000001-020 Account opening amount Opening amount float —

[0069] Table 3 Sample Data Table of Bank User Account Opening Information

[0070]

[0071]

[0072] Table 4 Data Interpretation Table

[0073]

[0074]

[0075] Using the large model, the system generates SQL statements based on the data recognition results in Table 4. Combined with the database write service, the system stores the recognition results as new columns in the original database table corresponding to the resource table, and splits the resource table into a field array according to the fields, such as... Figure 5 As shown;

[0076] The system iterates through the field array, uses RAG technology to find possible corresponding data elements for each field in the knowledge base, and then passes one or more of the recalled data elements into the context of the large model, allowing the large model to select appropriate data elements for comparison. Figure 6 As shown, the obtained data items are analyzed, the benchmarking data elements are extracted, and stored in the corresponding columns of the original database table. Finally, the results are passed into the large model context, and the benchmarking results of the resource table are output, as shown. Figure 7 As shown.

[0077] It should be noted that some industry data elements have strong business attributes. Directly benchmarking them by data element name alone is difficult to accurately reflect their complete business meaning. However, by leveraging the semantic understanding and analysis capabilities of large models, one or more explanatory understanding results can be automatically generated, improving the accuracy of benchmarking. For example, the "Contact Information" field in the "Employee Basic Information" table can also generate "Telephone Number, Mobile Number, and Communication Method" in combination with the context.

[0078] It should be noted that this application stores industry data elements in a vector knowledge base and relies on a scheduled task service mechanism to retrieve the resource table to be benchmarked from the database. Combined with the semantic understanding of the large model, automatic benchmarking is achieved. When the standards and benchmarking rules of cross-industry data elements are different, it is only necessary to replace the industry data elements in the knowledge base and modify the prompt words of the large model. There is no need to manually adjust the mapping relationship table or update the script, which reduces the maintenance burden. At the same time, this application improves the benchmarking accuracy based on the powerful semantic understanding capability of the large model.

[0079] It should be noted that this application realizes dynamic threshold adjustment, intelligent mapping generation, and rapid cross-industry adaptation. Compared with the traditional configuration file-based method, its core value lies in transforming manually maintained hard-coded rules into a learnable semantic matching model, enabling the system to have the advantages of adaptability, intelligence, and lightweight operation and maintenance. It is suitable for scenarios with inconsistent standards and complex heterogeneous systems in digital transformation, and can further assist in data governance and release data value.

[0080] The following is for reference. Figure 8 It shows a schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application. Figure 8 The electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0081] like Figure 8 As shown, the computer system includes a central processing unit (CPU) 801, which can perform various appropriate actions and processes based on programs stored in read-only memory (ROM) 802 or programs loaded from storage section 808 into random access memory (RAM) 803. RAM 803 also stores various programs and data required for system operation. CPU 801, ROM 802, and RAM 803 are interconnected via bus 804. Input / output (I / O) interface 805 is also connected to bus 804.

[0082] The following components are connected to I / O interface 805: an input section 806 including a keyboard, mouse, etc.; an output section 807 including a liquid crystal display (LCD) and speakers, etc.; a storage section 808 including a hard disk, etc.; and a communication section 809 including a network interface card such as a LAN card and a modem, etc. The communication section 809 performs communication processing via a network such as the Internet. A drive 810 is also connected to I / O interface 805 as needed. A removable medium 811, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., is installed on drive 810 as needed so that computer programs read from it can be installed into storage section 808 as needed.

[0083] Specifically, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable storage medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 809, and / or installed from removable medium 811. When the computer program is executed by central processing unit (CPU) 801, it performs the functions defined in the methods of this application. It should be noted that the computer-readable storage medium of this application can be a computer-readable signal medium or a computer-readable storage medium or any combination thereof. The computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to: electrical connections having one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination thereof. In this application, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can also be any computer-readable storage medium other than a computer-readable storage medium that can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. Program code contained on a computer-readable storage medium may be transmitted using any suitable medium, including but not limited to: wireless, wire, optical fiber, RF, etc., or any suitable combination thereof.

[0084] Computer program code for performing the operations of this application can be written in one or more programming languages ​​or a combination thereof. Programming languages ​​include object-oriented programming languages—such as Java, Smalltalk, and C++—as well as conventional procedural programming languages—such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0085] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this application. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0086] The modules described in the embodiments of this application can be implemented in software or in hardware.

[0087] In another aspect, this application also provides a computer-readable storage medium, which may be included in the electronic device described in the above embodiments; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable storage medium carries one or more programs. When the aforementioned one or more programs are executed by the electronic device, the electronic device causes the following: In response to a table name input by the user, the system transmits the metadata and sample data of the resource table corresponding to the table name to the large model; the large model generates semantic interpretations corresponding to each field in the resource table, obtaining recognition results; the system stores the recognition results as new columns in the original database table corresponding to the resource table, and splits the resource table into field arrays according to the fields; the system traverses the field arrays, using RAG technology to compare the elements in the field arrays with the data elements in the vector knowledge base, obtaining similar data elements in the field arrays; the large model selects similar data elements for benchmarking based on the metadata, obtaining the benchmarking results of the current resource table, and simultaneously stores the benchmarking results as new columns in the original database table.

[0088] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features with similar functions disclosed in this application.

[0089] Finally, it should be noted that the above embodiments are for illustration only and not for limiting the technical solutions of the present invention. Although the present invention has been described in detail with reference to the above embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the present invention without departing from the spirit and scope of the present invention. Any modifications or partial substitutions should be covered within the scope of the claims of the present invention.

Claims

1. An automated benchmarking method based on a large model, characterized in that, include: S1: In response to the table name entered by the user, the system passes the metadata and sample data of the resource table corresponding to the table name to the large model; S2: The large model generates semantic interpretations corresponding to each field in the resource table to obtain recognition results. The system stores the recognition results as new columns in the original database table corresponding to the resource table and splits the resource table into field arrays according to the fields. S3: The system traverses the field array and uses RAG technology to compare the elements in the field array with the data elements in the vector knowledge base to obtain similar data elements in the field array; as well as S4: The large model selects similar data elements for benchmarking based on the metadata, obtains the benchmarking result of the current resource table and outputs it, and at the same time stores the benchmarking result as a new column in the original database table.

2. The method according to claim 1, characterized in that, Prior to S1, standardized data elements and metadata for specific industries were stored in a vector knowledge base, with each data element treated as an independent vectorized segment.

3. The method according to claim 1, characterized in that, After the system passes the metadata and sample data of the corresponding resource table to the large model, if the parameters passed in the resource table are empty, the resource table does not exist and an exception is thrown.

4. The method according to claim 1, characterized in that, The RAG technology is used to compare the elements in the array with the data elements in the vector knowledge base, including: performing a similarity search in the vector knowledge base based on the field name, data type and the recognition result, and recalling one or more candidate data elements.

5. The method according to claim 4, characterized in that, It also includes inputting field information and its context, along with the recalled candidate data elements, into a large model, whereby the large model selects the most matching data element as the benchmark result for that field.

6. The method according to claim 1, characterized in that, The recognition results are automatically generated by the large model based on field names, sample data, and contextual semantics, and are used to expand the semantic representation of the fields, including synonyms, near-synonyms, or explanatory terms for each field.

7. A system for benchmarking using the method described in any one of claims 1-6, characterized in that, Includes a scheduled task scheduler and an automatic resource table matching module; The scheduled task scheduler is configured to perform timed matching of the resource tables of the automatic resource table matching module. The resource table automatic alignment module includes a metadata intelligent parsing module, a vector knowledge base, and a large model-assisted alignment module. The vector knowledge base is used to store standardized data elements and metadata for a specific industry. The metadata intelligent parsing module parses the metadata information in the vector knowledge base to obtain the parsing results. The large model-assisted benchmarking module uses the parsing results and, in conjunction with RAG technology, performs automated benchmarking from the original database tables to standardized data elements.

8. A computer program product having one or more computer programs thereon, characterized in that, When the computer program is executed by a computer processor, the method described in any one of claims 1-6 is performed.

Images