Report generation method, device and equipment based on large model and storage medium
By using automated processing based on large models, the problem of cumbersome business report generation has been solved, achieving efficient and reliable report generation while reducing manual intervention and time consumption.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- SHENZHEN TAIJI SOFTWARE CO LTD
- Filing Date
- 2025-12-23
- Publication Date
- 2026-04-17
AI Technical Summary
The existing business report generation process is cumbersome, consumes a lot of human resources and time, and results in low generation efficiency.
A large model-based approach is adopted. By acquiring task information uploaded by the client, the large model is used for semantic parsing and feature extraction to generate feature vectors. These vector vectors are then matched and queried in a vector database to automatically populate business templates and generate reports.
It reduces the time required to generate business reports, improves generation efficiency, enhances report reliability, and avoids errors and time consumption caused by manual intervention.
Smart Images
Figure CN121882006A_ABST
Abstract
Description
Technical Field
[0001] This application relates to the fields of artificial intelligence technology and data processing technology, and in particular to a method, apparatus, device and storage medium for generating reports based on large models. Background Technology
[0002] Against the backdrop of deepening digital transformation across industries, business reports have become a core basis for corporate decision-making, government regulation, and industry analysis. Business reports facilitate the systematic integration of scattered business indicator data.
[0003] However, the current process for generating business reports is cumbersome and hinders efficiency. This is because existing business reports primarily rely on manual import, requiring staff to expend significant time and effort manually collecting various data points from scattered and independent storage locations. This consumes substantial human and time resources, increasing the time required to generate business reports and thus hindering efficiency. Summary of the Invention
[0004] This application provides a report generation method, apparatus, device, and storage medium based on a large model to solve the technical problem that the existing business report generation process is cumbersome and not conducive to improving the efficiency of business report generation.
[0005] In a first aspect, embodiments of this application provide a report generation method based on a large model, applied to a server-side device, the report generation method comprising: Obtain task information uploaded by the client device, and extract the business identifier and business template from the task information. The business template includes multiple business data indicators to be populated. Multiple business data metrics are input into a large model, and the large model performs semantic parsing processing on each business data metric to obtain the semantic information of each business data metric. The semantic information of each business data indicator is input into the semantic feature extraction module. The semantic feature extraction module transforms and processes the semantic information of each business data indicator to generate a feature vector for each business data indicator. The business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item to obtain a mapping relationship table. Select multiple data items corresponding to successfully matched business data indicators as multiple target data items. Input the multiple target data items into the large model. The large model generates a query instruction for each target data item. Submit the query instruction for each target data item to the query interface of the vector database. Execute the query instruction for each target data item through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, import the business value stored in each target data item into the populate position of each successfully matched business data indicator in the business template to generate a business report.
[0006] In one possible implementation of the first aspect, the step of obtaining task information uploaded by the client device, and obtaining a business identifier and a business template from the task information, wherein the business template includes multiple business data indicators to be filled, including: Receive connection requests sent by client devices, the connection requests carrying the device identifier of the client devices; The device identifier of the client device is validated for legality. After the validity is validated, a data transmission channel is established between the server device and the client device. The task information uploaded by the client device is obtained through the data transmission channel, and the business identifier and business template are obtained from the task information.
[0007] In one possible implementation of the first aspect, the business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table, including: Using JSON format, the business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request, and the feature vector of each data item is obtained from the set of feature vectors. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is then established between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table.
[0008] In one possible implementation of the first aspect, the selection of multiple data items corresponding to successfully matched business data indicators as multiple target data items, inputting the multiple target data items into a large model, generating a query instruction for each target data item, submitting the query instruction for each target data item to the query interface of the vector database, executing the query instruction for each target data item through the query interface to obtain the business value stored in each target data item, and importing the business value stored in each target data item into the fill position of each successfully matched business data indicator in the business template according to the mapping relationship table, generating a business report, including: Select data items corresponding to multiple successfully matched business data indicators as multiple target data items, obtain prompt word templates, perform feature extraction on multiple target data items and prompt word templates, and obtain feature vectors of multiple target data items and feature vectors of prompt word templates; The feature vectors of multiple target data items and the feature vectors of prompt word templates are input into the large model. The large model generates a query instruction for each target data item and submits the query instruction for each target data item to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, the business value stored in each target data item is imported into the population position of each successfully matched business data indicator in the business template to generate a business report.
[0009] In one possible implementation of the first aspect, after selecting multiple data items corresponding to successfully matched business data indicators as multiple target data items, inputting these target data items into a large model, generating a query instruction for each target data item, submitting the query instruction for each target data item to the query interface of the vector database, executing the query instruction for each target data item through the query interface, obtaining the business value stored in each target data item, and importing the business value stored in each target data item into the fill position of each successfully matched business data indicator in the business template according to the mapping relationship table, and generating a business report, the business report generation method further includes: Send business reports to client devices via Hypertext Transfer Security Protocol.
[0010] In one possible implementation of the first aspect, the matching function is defined as follows: ; Formula variable explanation: Indicates the first Individual business data metrics; Indicates the first One data item; Indicates the first The dimensions of each data item; Indicates the first The dimensions of each data item; Indicates the first Atomic indices of each data item; Indicates the first Atomic indices of each data item; Indicates the first The first business data metric and the first Matching results between data items; : indicates the first The first business data metric and the first The matching result between the data items is a successful match; : indicates the first The first business data metric and the first The matching result between the data items is a failure.
[0011] In one possible implementation of the first aspect, the business templates include templates for monthly statistical business and templates for regional statistical business.
[0012] Secondly, embodiments of this application provide a report generation apparatus based on a large model, applied to a server-side device, comprising: The acquisition module is used to acquire task information uploaded by the client device, and to obtain the business identifier and business template from the task information. The business template includes multiple business data indicators to be filled in. The parsing module is used to input multiple business data metrics into a large model, and the large model performs semantic parsing processing on each business data metric to obtain the semantic information of each business data metric. The conversion module is used to input the semantic information of each business data indicator into the semantic feature extraction module, and the semantic feature extraction module converts the semantic information of each business data indicator to generate the feature vector of each business data indicator. The storage module is used to package the business identifier and data acquisition instruction to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item to obtain a mapping relationship table. The generation module selects multiple data items corresponding to successfully matched business data indicators as multiple target data items. It inputs these target data items into a large model, which generates a query instruction for each target data item. The query instruction for each target data item is then submitted to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. Based on the mapping relationship table, the business value stored in each target data item is imported into the populate position of each successfully matched business data indicator in the business template to generate a business report.
[0013] Thirdly, embodiments of this application provide a server device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the report generation method described in the first aspect above.
[0014] Fourthly, embodiments of this application provide a computer-readable storage medium storing a computer program that, when executed by a processor, implements the report generation method described in the first aspect above.
[0015] Fifthly, embodiments of this application provide a computer program product that, when run on a server device, causes the server device to execute the report generation method described in the first aspect.
[0016] The beneficial effects of the embodiments of this application are as follows: Firstly, multiple data items corresponding to successfully matched business data indicators are selected as multiple target data items. These target data items are input into a large model, which generates a query instruction for each target data item. The query instruction for each target data item is then submitted to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. Based on the mapping relationship table, the business value stored in each target data item is imported into the population position of each successfully matched business data indicator in the business template to generate a business report. Since there is no need to manually import the business value stored in each target data item, the generation time of the business report is reduced, which helps to improve the generation efficiency of the business report. Secondly, since business reports are automatically generated and not subject to human intervention, the reliability of business reports is improved. Attached Figure Description
[0017] To more clearly illustrate the technical solutions in the embodiments of this application, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0018] Figure 1 This is an application scenario diagram of the report generation method provided in the embodiments of this application; Figure 2 This is a flowchart illustrating the report generation method provided in an embodiment of this application; Figure 3 A flowchart illustrating the implementation of S205 provided in this application embodiment; Figure 4 A schematic block diagram of a report generation apparatus provided in the embodiments of this application; Figure 5 This is a schematic diagram of the structure of the server device provided in an embodiment of this application. Detailed Implementation
[0019] To make the objectives, technical solutions, and advantages of this application clearer, the following detailed description is provided in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the scope of this application. All other embodiments obtained by those skilled in the art based on the embodiments in this application without inventive effort are within the scope of protection of this application.
[0020] In the following description, specific details such as particular system architectures and techniques are set forth for illustrative purposes and not for limitation, in order to provide a thorough understanding of the embodiments of this application. However, those skilled in the art will understand that this application may also be implemented in other embodiments without these specific details. In other instances, detailed descriptions of well-known systems, apparatuses, circuits, and methods have been omitted so as not to obscure the description of this application with unnecessary detail.
[0021] It should be understood that in the description of this application and the appended claims, the terms "first," "second," "third," etc., are used only to distinguish descriptions and should not be construed as indicating or implying relative importance. The terms "comprising," "including," "having," and their variations all mean "including but not limited to," unless otherwise specifically emphasized.
[0022] Furthermore, the technical solutions of the various embodiments can be combined with each other, but only if they are based on the ability of those skilled in the art to implement them. When the combination of technical solutions is contradictory or cannot be implemented, it should be considered that such combination of technical solutions does not exist and is not within the scope of protection claimed in this application.
[0023] The flowchart shown in the attached diagram is for illustrative purposes only and does not necessarily include all content and operations / steps, nor does it necessarily have to be performed in the described order. For example, some operations / steps can be broken down, combined, or partially merged, so the actual execution order may change depending on the actual situation.
[0024] The report generation method provided in this application embodiment can be applied to servers, mobile phones, tablets, wearable devices, in-vehicle devices, laptops, and ultra-mobile personal computers (UMPCs). This application embodiment does not impose any restrictions on the specific type of server-side device.
[0025] Please see Figure 1 , Figure 1 The application scenario diagram of the report generation method provided in the embodiments of this application is described in detail below: The server device receives a connection request sent by the client device, and the connection request carries the device identifier of the client device; The server device verifies the validity of the client device's device identifier. Once the validity verification is successful, a data transmission channel is established between the server device and the client device. The server device then obtains the task information uploaded by the client device through the data transmission channel and extracts the business identifier and business template from the task information.
[0026] In this embodiment, the server device verifies the validity of the client device's device identifier. Once the validity verification is successful, a data transmission channel is established between the server device and the client device, thereby reducing service congestion caused by the influx of unauthorized devices.
[0027] Please see Figure 2 , Figure 2 This is a flowchart illustrating the report generation method provided in this application embodiment, which can be applied to server-side devices.
[0028] like Figure 2 As shown in the embodiments of this application, the report generation method includes the following steps, which are detailed below: S201, Obtain task information uploaded by the client device, and retrieve the business identifier and business template from the task information. The business template includes multiple business data indicators to be filled in. The business templates include templates for monthly statistical business and templates for regional statistical business.
[0029] S202: Input multiple business data metrics into the large model, and perform semantic parsing processing on each business data metric through the large model to obtain the semantic information of each business data metric. Among them, large-scale pre-trained models are a highly anticipated technological achievement in the field of artificial intelligence. They use massive amounts of data as learning materials, which come from a wide range of sources, including text, images, audio, and other forms on the Internet, and often reach hundreds of millions or even billions of data points.
[0030] S203, input the semantic information of each business data indicator into the semantic feature extraction module, and transform the semantic information of each business data indicator through the semantic feature extraction module to generate the feature vector of each business data indicator. S204: Package the business identifier and data acquisition instruction to obtain a data acquisition request. Transmit the data acquisition request to the access interface of the vector database. Obtain the feature vector set returned by the access interface according to the data acquisition request. Obtain the feature vector of each data item from the feature vector set. Use a matching function to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. Establish a mapping relationship between each successfully matched business data indicator and the data item corresponding to each successfully matched business data indicator to obtain a mapping relationship table. Specifically, the process involves packaging the business identifier and data acquisition instruction to obtain a data acquisition request, transmitting the data acquisition request to the access interface of the vector database, obtaining the feature vector set returned by the access interface based on the data acquisition request, retrieving the feature vector of each data item from the feature vector set, using a matching function to match the feature vector of each business data indicator with the feature vector of each data item in the vector database, obtaining each successfully matched business data indicator, and establishing a mapping relationship between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table, including: Using JSON format, the business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request, and the feature vector of each data item is obtained from the set of feature vectors. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is then established between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table.
[0031] The matching function is defined as follows: ; Formula variable explanation: Indicates the first Individual business data metrics; Indicates the first One data item; Indicates the first The dimensions of each data item; Indicates the first The dimensions of each data item; Indicates the first Atomic indices of each data item; Indicates the first Atomic indices of each data item; Indicates the first The first business data metric and the first Matching results between data items; : indicates the first The first business data metric and the first The matching result between the data items is a successful match; : indicates the first The first business data metric and the first The matching result between the data items is a failure.
[0032] In this method, the feature vector of each business data indicator is matched with the feature vector of each data item in the vector database. This effectively avoids the matching errors caused by factors such as differences in literal expression, synonymous heterogeneous expressions, and noise interference in traditional solutions, and greatly reduces the probability of mismatch and missed match.
[0033] S205: Select multiple data items corresponding to successfully matched business data indicators as multiple target data items, input the multiple target data items into the large model, the large model generates a query instruction for each target data item, submit the query instruction for each target data item to the query interface of the vector database, execute the query instruction for each target data item through the query interface, obtain the business value stored in each target data item, and import the business value stored in each target data item into the fill position of each successfully matched business data indicator in the business template according to the mapping relationship table, and generate a business report.
[0034] Specifically, the business values stored for each target data item are imported into the populate positions of each successfully matched business data metric in the business template to generate a business report, including: The business values stored for each target data item are imported into the populate position of each successfully matched business data indicator in the business template. The format of the populated business template is then validated. Once the format validation is passed, the populated business template is used as the business report.
[0035] For example, if the business template after filling is an Excel template, the formulas in the business template after filling are validated to ensure the correctness of the formula calculation results after the data is filled. For example, if the filled business template is a Word template, the font, font size, and alignment of the filled business template are adjusted to maintain the original style of the business template.
[0036] For example, when the business template after filling is a PDF template, the text position of the filled business template is precisely adjusted to avoid data overlap or misalignment.
[0037] Specifically, based on the mapping relationship table, the business values stored in each target data item are imported into the fill position of each successfully matched business data indicator in the business template to generate a business report. The mapping relationship table clearly specifies the business values stored in each target data item, avoiding problems such as incorrect value placement and format incompatibility caused by misunderstandings during manual filling. This ensures the accuracy of the business report and shortens the business report production cycle.
[0038] The business templates include templates for monthly statistical business and templates for regional statistical business.
[0039] For ease of explanation, let's take a business template that includes monthly statistics as an example, as shown below: For example, there are three target data items: sales revenue, year-on-year growth rate, and employment rate. The statistics for October 2025 are: sales revenue of RMB 12.5689 million, year-on-year growth rate of 8.3%, and employment rate of 96.5%. Enter the above data in the fill-in positions of the three business data indicators respectively. The data display area of the business report will show: sales revenue in October 2025: RMB 12.5689 million; sales revenue year-on-year growth rate: 8.3%; employment rate: 96.5%.
[0040] For ease of explanation, let's take a regional statistics business template as an example, as shown below: For example, there are three target data items: sales revenue, year-on-year growth rate, and employment rate. If the statistical data for a certain high-tech zone in October 2025 are: enterprise sales revenue of 865.205 million yuan, year-on-year growth rate of enterprise sales revenue of 6.7%, and employment rate of the population of the zone of 94.2%, then fill in the above data in the corresponding fill-in spaces. After completion, the data display area of the business report will show: Enterprise sales revenue of the high-tech zone in October 2025: 865.205 million yuan; year-on-year growth rate: 6.7%; employment rate: 94.2%. The process involves selecting multiple data items corresponding to successfully matched business data indicators as multiple target data items, inputting these target data items into a large model, generating a query instruction for each target data item, submitting this query instruction to the query interface of the vector database, executing the query instruction through the query interface to obtain the stored business value for each target data item, and importing the stored business value for each target data item into the population position of each successfully matched business data indicator in the business template according to the mapping relationship table. After generating the business report, the business report generation method also includes: Send business reports to client devices via Hypertext Transfer Security Protocol.
[0041] Hypertext Transfer Protocol Secure (HTTPS) is a transmission protocol for secure communication over computer networks.
[0042] The beneficial effects of the embodiments of this application are as follows: Firstly, multiple data items corresponding to successfully matched business data indicators are selected as multiple target data items. These target data items are input into a large model, which generates a query instruction for each target data item. The query instruction for each target data item is then submitted to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. Based on the mapping relationship table, the business value stored in each target data item is imported into the population position of each successfully matched business data indicator in the business template to generate a business report. Since there is no need to manually import the business value stored in each target data item, the generation time of the business report is reduced, which helps to improve the generation efficiency of the business report. Secondly, since business reports are automatically generated and not subject to human intervention, the reliability of business reports is improved.
[0043] Please see Figure 3 , Figure 3 The implementation flowchart of S205 provided in the embodiments of this application is described in detail below: S301, Select data items corresponding to multiple successfully matched business data indicators as multiple target data items, obtain prompt word templates, perform feature extraction on multiple target data items and prompt word templates, and obtain feature vectors of multiple target data items and feature vectors of prompt word templates; S302, input the feature vectors of multiple target data items and the feature vectors of prompt word templates into the large model. The large model generates a query instruction for each target data item and submits the query instruction for each target data item to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, the business value stored in each target data item is imported into the fill position of each successfully matched business data indicator in the business template to generate a business report.
[0044] In this embodiment of the application, the business value stored in each target data item is imported into the filling position of each successfully matched business data indicator in the business template according to the mapping relationship table, and a business report is generated. This avoids the omissions and time-consuming process that may occur when manually entering each item, and greatly shortens the generation cycle of the business report, thus improving the generation efficiency of the business report.
[0045] For the report generation method described in the above embodiments, please refer to [link / reference]. Figure 4 , Figure 4 This is a schematic block diagram of a report generation apparatus provided in an embodiment of this application. Figure 4 The report generation device 400 shown can be applied to, for example... Figure 1 The application scenario diagram shows the server-side device. The following section uses the server-side device as an example to illustrate this. Figure 4 The report generation device 400 shown will be described in detail. The report generation device 400 may include an acquisition module 401, a parsing module 402, a conversion module 403, a saving module 404, and a generation module 405.
[0046] The acquisition module 401 is used to acquire task information uploaded by the client device, and to obtain the business identifier and business template from the task information. The business template includes multiple business data indicators to be filled in. The parsing module 402 is used to input multiple business data indicators into the large model, and to perform semantic parsing processing on each business data indicator through the large model to obtain the semantic information of each business data indicator. The conversion module 403 is used to input the semantic information of each business data indicator into the semantic feature extraction module, and to convert the semantic information of each business data indicator through the semantic feature extraction module to generate the feature vector of each business data indicator. The storage module 404 is used to package the business identifier and data acquisition instruction to obtain a data acquisition request, transmit the data acquisition request to the access interface of the vector database, obtain the feature vector set returned by the access interface according to the data acquisition request, obtain the feature vector of each data item from the feature vector set, use a matching function to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator, establish a mapping relationship between each successfully matched business data indicator and the data item corresponding to each successfully matched business data indicator, and obtain a mapping relationship table. The generation module 405 is used to select multiple data items corresponding to successfully matched business data indicators as multiple target data items, input the multiple target data items into the large model, generate a query instruction for each target data item, submit the query instruction for each target data item to the query interface of the vector database, execute the query instruction for each target data item through the query interface, obtain the business value stored in each target data item, and import the business value stored in each target data item into the fill position of each successfully matched business data indicator in the business template according to the mapping relationship table, and generate a business report.
[0047] It should be noted that the various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. The same or similar parts between the various embodiments can be referred to each other.
[0048] The beneficial effects of the embodiments of this application are as follows: Firstly, multiple data items corresponding to successfully matched business data indicators are selected as multiple target data items. These target data items are input into a large model, which generates a query instruction for each target data item. The query instruction for each target data item is then submitted to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. Based on the mapping relationship table, the business value stored in each target data item is imported into the population position of each successfully matched business data indicator in the business template to generate a business report. Since there is no need to manually import the business value stored in each target data item, the generation time of the business report is reduced, which helps to improve the generation efficiency of the business report. Secondly, since business reports are automatically generated and not subject to human intervention, the reliability of business reports is improved.
[0049] Please see Figure 5 , Figure 5 This is a schematic diagram of the structure of the server device provided in an embodiment of this application.
[0050] like Figure 5 As shown, Figure 5 The server device 2 includes: at least one processor 20, a memory 21, and a computer program 22 stored in the memory 21 and executable on the at least one processor 20, wherein the processor 20 executes the computer program 22 to implement the steps in any of the above method embodiments.
[0051] The server-side device 2 may include, but is not limited to, a processor 20 and a memory 21. Those skilled in the art will understand that... Figure 5 This is merely an example of server device 2 and does not constitute a limitation on server device 2. It may include more or fewer components than shown in the figure, or combine certain components, or different components, such as input / output devices, network access devices, etc.
[0052] The processor 20 is used to run a computer program 22 stored in the memory 21, and performs the following steps when executing the computer program 22: Obtain task information uploaded by the client device, and extract the business identifier and business template from the task information. The business template includes multiple business data indicators to be populated. Multiple business data metrics are input into a large model, and the large model performs semantic parsing processing on each business data metric to obtain the semantic information of each business data metric. The semantic information of each business data indicator is input into the semantic feature extraction module. The semantic feature extraction module transforms and processes the semantic information of each business data indicator to generate a feature vector for each business data indicator. The business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item to obtain a mapping relationship table. Select multiple data items corresponding to successfully matched business data indicators as multiple target data items. Input the multiple target data items into the large model. The large model generates a query instruction for each target data item. Submit the query instruction for each target data item to the query interface of the vector database. Execute the query instruction for each target data item through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, import the business value stored in each target data item into the populate position of each successfully matched business data indicator in the business template to generate a business report.
[0053] The processor 20 may be a Central Processing Unit (CPU), or it may be other general-purpose processors, digital signal processors, field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. A general-purpose processor may be a microprocessor or any conventional processor.
[0054] In some embodiments, the memory 21 may be an internal storage unit of the server device 2, such as a hard disk or memory of the server device 2. In other embodiments, the memory 21 may be an external storage device of the server device 2, such as a plug-in hard disk, smart media card (SMC), secure digital (SD) card, flash card, etc. equipped on the server device 2.
[0055] Furthermore, the memory 21 may include both internal storage units of the server device 2 and external storage devices. The memory 21 is used to store the operating system, applications, boot loader, data, and other programs, such as the program code of the computer program. The memory 21 can also be used to temporarily store data that has been output or will be output.
[0056] It should be noted that the information interaction and execution process between the above-mentioned devices / units are based on the same concept as the method embodiments of this application. For details on their specific functions and technical effects, please refer to the method embodiments section, and they will not be repeated here.
[0057] This application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the steps described in the various method embodiments above.
[0058] The computer-readable storage medium may also be an external storage device of the report generation device or server equipment, such as a plug-in hard drive, smart media card (SMC), secure digital (SD) card, flash card, or non-transitory computer-readable storage medium equipped on the report generation device or server equipment.
[0059] Since the computer program stored in the computer-readable storage medium can execute any of the large model-based report generation methods provided in the embodiments of this application, the computer-readable storage medium can achieve the beneficial effects that any of the large model-based report generation methods provided in the embodiments of this application can achieve, as detailed in the preceding embodiments, and will not be repeated here.
[0060] This application provides a computer program product that, when run on a server device, causes the server device to execute the aforementioned report generation method.
[0061] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium.
[0062] Based on this understanding, all or part of the processes in the methods of the above embodiments of this application can be implemented by a computer program instructing related hardware. The computer program can be stored in a computer-readable storage medium, and when executed by a processor, it can implement the steps of the various method embodiments described above. The computer program includes computer program code, which can be in the form of source code, object code, executable files, or certain intermediate forms. The computer-readable medium includes: an entity or device for carrying computer program code to a server device, a recording medium, a computer memory, a read-only memory (ROM), a random access memory (RAM), an electrical carrier signal, a telecommunication signal, and a software distribution medium.
[0063] In the above embodiments, the descriptions of each embodiment have different focuses. For parts that are not described in detail or recorded in a certain embodiment, please refer to the relevant descriptions of other embodiments.
[0064] The above are merely preferred embodiments of this application and do not limit the patent scope of this application. Any equivalent structural or procedural transformations made using the content of this application's specification and drawings, or direct or indirect applications in other related technical fields, are similarly included within the patent protection scope of this application.
Claims
1. A report generation method based on a large model, characterized in that, Applied to server-side devices, the report generation method includes: Obtain task information uploaded by the client device, and extract the business identifier and business template from the task information. The business template includes multiple business data indicators to be populated. Multiple business data metrics are input into a large model, and the large model performs semantic parsing processing on each business data metric to obtain the semantic information of each business data metric. The semantic information of each business data indicator is input into the semantic feature extraction module. The semantic feature extraction module transforms and processes the semantic information of each business data indicator to generate a feature vector for each business data indicator. The business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item to obtain a mapping relationship table. Select multiple data items corresponding to successfully matched business data indicators as multiple target data items. Input the multiple target data items into the large model. The large model generates a query instruction for each target data item. Submit the query instruction for each target data item to the query interface of the vector database. Execute the query instruction for each target data item through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, import the business value stored in each target data item into the populate position of each successfully matched business data indicator in the business template to generate a business report.
2. The business report generation method according to claim 1, characterized in that, The step involves obtaining task information uploaded by the client device, retrieving a business identifier and a business template from the task information, and the business template including multiple business data indicators to be filled in, including: Receive connection requests sent by client devices, the connection requests carrying the device identifier of the client devices; The device identifier of the client device is validated for legality. After the validity is validated, a data transmission channel is established between the server device and the client device. The task information uploaded by the client device is obtained through the data transmission channel, and the business identifier and business template are obtained from the task information.
3. The business report generation method according to claim 1, characterized in that, The process involves packaging the business identifier and data acquisition instruction to obtain a data acquisition request, transmitting the data acquisition request to the access interface of the vector database, obtaining the feature vector set returned by the access interface based on the data acquisition request, retrieving the feature vector of each data item from the feature vector set, using a matching function to match the feature vector of each business data indicator with the feature vector of each data item in the vector database, obtaining each successfully matched business data indicator, and establishing a mapping relationship between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table, including: Using JSON format, the business identifier and data acquisition instruction are packaged to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request, and the feature vector of each data item is obtained from the set of feature vectors. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is then established between each successfully matched business data indicator and the corresponding data item, resulting in a mapping relationship table.
4. The business report generation method according to claim 1, characterized in that, The process involves selecting multiple data items corresponding to successfully matched business data indicators as multiple target data items, inputting these target data items into a large model, generating a query instruction for each target data item, submitting this query instruction to the query interface of the vector database, executing the query instruction through the query interface to obtain the business value stored in each target data item, and importing the business value stored in each target data item into the population position of each successfully matched business data indicator in the business template according to the mapping relationship table, thereby generating a business report, including: Select data items corresponding to multiple successfully matched business data indicators as multiple target data items, obtain prompt word templates, perform feature extraction on multiple target data items and prompt word templates, and obtain feature vectors of multiple target data items and feature vectors of prompt word templates; The feature vectors of multiple target data items and the feature vectors of prompt word templates are input into the large model. The large model generates a query instruction for each target data item and submits the query instruction for each target data item to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. According to the mapping relationship table, the business value stored in each target data item is imported into the population position of each successfully matched business data indicator in the business template to generate a business report.
5. The business report generation method according to claim 1, characterized in that, After selecting multiple data items corresponding to successfully matched business data indicators as multiple target data items, these target data items are input into a large model. The large model generates a query instruction for each target data item, which is then submitted to the query interface of the vector database. The query interface executes the query instruction for each target data item, obtaining the stored business value for each target data item. Based on the mapping table, the stored business value for each target data item is imported into the populate position of each successfully matched business data indicator in the business template. After generating the business report, the business report generation method also includes: Send business reports to client devices via Hypertext Transfer Security Protocol.
6. The business report generation method according to claim 1, characterized in that, The matching function is defined as follows: ； Formula variable explanation: Indicates the first Individual business data metrics; Indicates the first One data item; Indicates the first The dimensions of each data item; Indicates the first The dimensions of each data item; Indicates the first Atomic indices of each data item; Indicates the first Atomic indices of each data item; Indicates the first The first business data metric and the first Matching results between data items; : indicates the first The first business data metric and the first The matching result between the data items is a successful match; : indicates the first The first business data metric and the first The matching result between the data items is a failure.
7. The business report generation method according to claim 1, characterized in that, The business templates include templates for monthly statistical business and templates for regional statistical business.
8. A report generation device based on a large model, characterized in that, Applied to server-side devices, including: The acquisition module is used to acquire task information uploaded by the client device, and to obtain the business identifier and business template from the task information. The business template includes multiple business data indicators to be filled in. The parsing module is used to input multiple business data metrics into a large model, and the large model performs semantic parsing processing on each business data metric to obtain the semantic information of each business data metric. The conversion module is used to input the semantic information of each business data indicator into the semantic feature extraction module, and the semantic feature extraction module converts the semantic information of each business data indicator to generate the feature vector of each business data indicator. The storage module is used to package the business identifier and data acquisition instruction to obtain a data acquisition request. The data acquisition request is then transmitted to the access interface of the vector database. The access interface returns a set of feature vectors based on the data acquisition request. The feature vector of each data item is obtained from the feature vector set. A matching function is used to match the feature vector of each business data indicator with the feature vector of each data item in the vector database to obtain each successfully matched business data indicator. A mapping relationship is established between each successfully matched business data indicator and the corresponding data item to obtain a mapping relationship table. The generation module selects multiple data items corresponding to successfully matched business data indicators as multiple target data items. It inputs these target data items into a large model, which generates a query instruction for each target data item. The query instruction for each target data item is then submitted to the query interface of the vector database. The query instruction for each target data item is executed through the query interface to obtain the business value stored in each target data item. Based on the mapping relationship table, the business value stored in each target data item is imported into the populate position of each successfully matched business data indicator in the business template to generate a business report.
9. A server-side device, comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the computer program, it implements the report generation method as described in any one of claims 1 to 7.
10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by a processor, it implements the report generation method as described in any one of claims 1 to 7.