Project exhaustion report automatic generation method based on multi-source data integration

Abstract

The invention relates to the technical field of data integration, and discloses a multi-source data integration-based project exhaustion report automatic generation method. According to the multi-source data integration-based project exhaustion report automatic generation method, a report template is preset, public content is defined as fixed content, specific content is set as variables and numbered, and then required information is obtained from different data sources and comprises the steps of obtaining enterprise information from an API interface, extracting pledge information from a network through an RPA technology, and generating a project exhaustion report. And identifying the content of the contract file by using an OCR technology, removing repeated data, numbering, matching variable numbers in a preset template with the obtained data numbers to form a mapping relationship, finally integrating fixed contents and variable contents according to the preset template, generating a feasibility research report in a uniform format, and performing report consistency check to obtain a feasibility research report. The accuracy of the generation time and the report is evaluated, the generation efficiency and accuracy of the exhaustion report are improved, and the requirements of different projects are met.

Description

Technical Field

[0001] The present invention relates to the technical field of data integration, and specifically provides an automated generation method for project due diligence reports based on multi-source data integration. Background Art

[0002] When Puhui Company submits a non-financial business for approval, a project feasibility study report needs to be formed. The content of the feasibility study report generally includes: introduction of the applicant, description of the engineering project, guarantee plan, risk points and countermeasures, declaration of special matters, conclusion, etc. The feasibility study report is the most important reference material in the business approval process. Therefore, business personnel must ensure the accuracy of the content in the feasibility study report. Currently, business personnel produce the feasibility study report through manual editing, and there are the following problems: 1) The report content is extensive, and the manual editing method is inefficient, resulting in a large workload for business personnel; 2) The report data comes from numerous sources, such as big data systems, China Trustee Association website, contract documents, etc. Omissions are likely to occur through the manual copy + paste method; 3) The report data is not structured, and the reports produced manually have inconsistent formats, affecting the approval efficiency. Summary of the Invention

[0003] (1) Technical Problems to be Solved

[0004] In view of the deficiencies of the prior art, the present invention provides an automated generation method for project due diligence reports based on multi-source data integration. By various technical methods, the required data is obtained from each data source, and the data is structured. A feasibility study report with a unified format is automatically generated according to a preset report template, solving the problems of low efficiency, error-proneness, and inconsistent formats in the current manual method.

[0005] (2) Technical Solutions

[0006] To achieve the above object, the present invention provides the following technical solutions: An automated generation method for project due diligence reports based on multi-source data integration, including the following steps:

[0007] S1. Preset a report template. For reports of different projects, define the common content as fixed content, the specific content as variables, and number the variables.

[0008] S2. After obtaining the required data from each data source through different methods, remove duplicate data and number the data, including obtaining enterprise information from the big data platform through an interface, obtaining mortgage information from the China Trustee Association website through RPA, and obtaining project information by OCR recognizing the content of contract documents.

[0009] S3. Match the variable numbers defined in S1 with the data numbers obtained in S2 to form a mapping relationship.

[0010] S4. Integrate the fixed content and variable content according to the preset template to form a feasibility study report in a unified format, and conduct a report consistency check to evaluate the generation time of the preset template and the accuracy rate of the feasibility study report.

[0011] Preferably, the formula for the variable number is as follows:

[0012] V n = V base + n

[0013] In the formula, V n represents the number of the nth variable, V base represents the initial value of the variable number, and n represents the serial number of the variable.

[0014] Preferably, the formula for data acquisition is as follows:

[0015] D i = f(S i,j , T k )

[0016] In the formula, D i represents the data obtained from the ith data source, S i,j represents the jth type of information in the ith data source, and T k represents the data acquisition method.

[0017] Preferably, the formula for removing duplicate data is as follows:

[0018] D unique = D - D dup

[0019] In the formula, D unique represents the valid data retained after the deduplication operation, D represents all the originally obtained data items, and D dup represents the set of duplicate data.

[0020] Preferably, the formula for numbering the data is as follows:

[0021] N k = D source (k)

[0022] In the formula, N k represents the number of the kth obtained data, and D source (k) represents the kth information extracted from the data source.

[0023] Preferably, the mapping relationship is as follows:

[0024] M(V n , N k ) → (matched)

[0025] In the formula, M represents the mapping relationship function, V n represents the variable number defined in the report template, N represents the corresponding data number obtained from the data source, and matched represents the returned matching result.

[0026] Preferably, the integration formula of the fixed content and the variable content is as follows:

[0027] R = C f + C v

[0028] In the formula, R represents the finally generated report result, C f represents the fixed content in the report template, and C v represents the variable content filled according to the mapping relationship.

[0029] Preferably, the formula for the report consistency check is as follows:

[0030] Check(R) = {R min , R max , R avg}

[0031] In the formula, Check(R) represents the consistency check function, and R min represents the minimum value of the data in the report, R max represents the maximum value of the data in the report, and R avg represents the average value of the data in the report.

[0032] Preferably, the formula for evaluating the generation time of the preset template is as follows:

[0033]

[0034] In the formula, T g represents the total time required to generate the template, T j represents the time required to generate each part of the template, n represents the different parts combined in the report, and j represents the index subscript.

[0035] Preferably, the formula for evaluating the accuracy rate of the feasibility study report is as follows:

[0036]

[0037] In the formula, Acc represents the accuracy rate of the feasibility study report, and C correct represents the number of correctly matched contents, and C total represents the number of all contents in the report, including the matched and unmatched contents.

[0038] Compared with the prior art, the present invention provides a method for automatically generating a project due diligence report based on multi-source data integration, which has the following beneficial effects:

[0039] 1. By structuring the report content and unifying the report format, the present invention makes the risk control information clear at a glance and improves the business approval efficiency.

[0040] 2. By automatically obtaining the required data by the system and generating a report according to a fixed format, the present invention can not only avoid omissions in the content but also reduce the workload of business personnel. BRIEF DESCRIPTION OF THE DRAWINGS

[0041] Figure 1 It is a schematic diagram of the method steps of the present invention. DETAILED DESCRIPTION OF THE INVENTION

[0042] The following will clearly and completely describe the technical solutions in the embodiments of the present invention with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0043] In view of the problems that currently business personnel make feasibility study reports in a manual editing manner, there are many report contents, the manual editing method is inefficient, resulting in a large workload for business personnel, there are many sources of report data, it is easy to produce omissions through the way of manual copying and pasting, and the report data is not structured, and the reports made by manual methods have inconsistent formats, affecting the approval efficiency. For this reason, a method for automatically generating a project due diligence report based on multi-source data integration is proposed. Please refer to Figure 1 , and the method includes the following steps:

[0044] S1. Preset a report template. For reports of different projects, define the common content as fixed content, the specific content as variables, and number the variables;

[0045] In the method for automatically generating a project due diligence report, presetting a report template is a fundamental and key step. By defining the structure of the report, the common content and the specific content can be effectively distinguished, thereby improving the writing efficiency and consistency of the report. In this process, the common content (such as the basic information of the enterprise, historical performance, market environment, etc.) is defined as fixed content, while the specific content is defined as variables. At this time, variable numbering becomes an indispensable part to ensure the uniqueness of these variables in the report. Specifically, the formula for variable numbering is:

[0046] V n =V base +n

[0047] In this formula, V n represents the number of the nth variable. V base is the base number, usually starting from 1. n represents the serial number of the variable. In this way, the team can assign a clear and easily recognizable number to each specific content;

[0048] The benefits of this approach are numerous and very specific. First of all, variable numbering provides a systematic method to manage and reference specific content. When writing reports, team members don't have to worry about variable duplication or confusion. Each number is directly associated with specific project data. Secondly, this structured management method greatly simplifies the process of variable filling. The system can quickly match the variable numbers in the template with the data in the actual data source, ensuring that the data is accurately filled into the appropriate position in the report, thus achieving the consistency of data and content. In addition, this method also greatly improves the team's collaboration efficiency. In a multi-person cooperation environment, the preset report template and its clear variable numbers reduce the communication cost, ensuring that everyone can quickly and accurately understand the content and corresponding data that need to be filled;

[0049] Finally, when dynamically adjusting and updating reports, using variable numbers can quickly identify the parts that need to be modified without having to comb through the entire report, truly achieving efficient and flexible report generation and management. This not only improves the work quality but also provides the team with an efficient tool, enabling them to focus on analysis and decision-making rather than tedious data processing;

[0050] S2. After obtaining the required data from various data sources in different ways, remove duplicate data and perform data numbering, including obtaining enterprise information from the big data platform through an interface, obtaining mortgage information from the China Bond Depository and Clearing Corporation Limited website through RPA, and obtaining project information by identifying the content of contract documents through OCR;

[0051] In the process of automatic generation of project due diligence reports, by adopting a variety of advanced technical means, efficient information extraction can be carried out from different data sources. First of all, using an interface to obtain enterprise information from the big data platform can quickly obtain key data such as company size, financial status, and market share. This information is crucial for evaluating the potential value of the project. Secondly, by using robotic process automation (RPA) to extract mortgage information from the China Bond Depository and Clearing Corporation Limited website, complex data can be automatically and accurately obtained, reducing errors caused by manual operations. And by using optical character recognition (OCR) technology to identify and extract the content of contract documents, the data in paper documents can be effectively converted into an operable digital format. The comprehensive application of these three technical means ensures efficient and accurate information acquisition in the face of the diversity and complexity of data sources;

[0052] After data acquisition, the system needs to remove duplicate data to ensure the establishment of a clear and concise dataset. At this time, the following formula can be applied:

[0053] D unique = D - D dup

[0054] Where D unique represents the unique data set, D is the total data set, including all the acquired data, and D dup is the duplicate data set. By identifying and removing duplicates, the system can ensure the validity and accuracy of the data;

[0055] After deduplication, each data item will be independently numbered for use in subsequent analysis and report generation. For the process of data numbering, the formula can be used:

[0056] N k = D source (k)

[0057] Here, N k represents the number of the k-th data, and D source (k) represents the k-th piece of information obtained from the data source. By numbering the data, each piece of information can be clearly referenced and retrieved, which is particularly important in team collaboration;

[0058] The advantage of this technical means is that it greatly improves the efficiency of data processing, can extract and clean accurate information from a large amount of data in a short time. At the same time, removing duplicate data and accurate numbering make the data more reliable and consistent in subsequent use, reducing the risks brought by manual operations. And due to the automation of the data acquisition and processing process, the team can focus on data analysis and decision-making rather than heavy manual sorting work. This efficient workflow not only improves the speed and quality of project due diligence, but also lays a solid data foundation for subsequent business expansion;

[0059] S3. By matching the variable numbers defined in S1 with the numbers of the data obtained in S2, a mapping relationship is formed;

[0060] In the process of automatic generation of project due diligence reports, data is a core asset. By matching the variable numbers defined in S1 with the numbers of the data obtained in S2, the system can form a clear and concise mapping relationship. This process not only improves the accuracy and systematicness of information processing, but also makes report generation more efficient. After defining these two numbers, the system performs matching through the mapping relationship formula:

[0061] M(V n ,N k)→(matched)

[0062] This formula describes how to map the variable numbers defined in the report template to the data numbers obtained from the data source, establishing a correlation between the two. The formation of the mapping relationship enables the accurate filling of data into the corresponding report entries;

[0063] The advantage of this technical approach is that it significantly improves the efficiency of data integration. Since the system can automatically match variables with data, it greatly reduces the time and workload of manual input and proofreading. By introducing a clear mapping relationship, the team can ensure the accuracy of the report content and avoid the citation of incorrect data. In a dynamically changing business environment, this accuracy is particularly important as it helps decision-makers rely on the latest and most relevant data for judgment;

[0064] In addition, through the visualization of the mapping relationship, team members can more intuitively understand the process of data flow and content generation, enhancing the transparency of team collaboration and the efficiency of communication. This not only improves the quality and speed of report generation but also provides more comprehensive and accurate decision-making support for the enterprise;

[0065] S4. Integrate the fixed content and variable content according to the preset template to form a feasibility study report in a unified format, and conduct a report consistency check to evaluate the generation time of the preset template and the accuracy rate of the feasibility study report;

[0066] After completing the data collection and collation of the project due diligence report, integrating the fixed content and variable content according to the preset template is a crucial step. This process not only ensures the consistency of the report format but also provides a systematic information presentation in the entire project evaluation. In this link, first use the formula for integrating the fixed content and variable content:

[0067] R = C f +C v

[0068] where R represents the finally generated feasibility study report, and C f is the fixed content, and this part of the information remains unchanged in all reports, such as the enterprise background, market analysis, etc. C v is the specific variable content, which varies according to each project. Through such integration, the report not only maintains a consistent format but also effectively conveys the information of a specific project, thus meeting the needs of decision-makers;

[0069] After the report is generated, conducting a consistency check is an important link to ensure data accuracy. By using the report consistency check formula:

[0070] Check(R) = {R min ,R max,R avg}

[0071] Here, Check(R) is a function for consistency checking, and R min , R max and R avg represent the minimum, maximum, and average values of the data in the report respectively. This kind of check can effectively identify data anomalies or inconsistencies, ensure the comprehensiveness and accuracy of report analysis, and reduce the risks brought by information errors;

[0072] In addition to data consistency checking, evaluating the generation time of the preset template is also an important step. Use the formula for the template generation time:

[0073]

[0074] Here, T g represents the total time required to generate the entire template, while T j is the time required to generate each partial template. By evaluating the generation time, the team can understand the time required for each step, thereby identifying and optimizing possible bottlenecks and improving the overall efficiency of report generation;

[0075] Finally, the accuracy of the report is crucial for the effectiveness of decision-making. Use the formula for calculating the accuracy of the report:

[0076]

[0077] Here, Acc is the accuracy of the report, C correct represents the number of correctly matched contents, while C total is the number of all contents in the report. Ensuring the accuracy of the report not only enhances the credibility of the report but also enables decision-makers to have more confidence in the obtained insight information, promoting the successful implementation of key decisions;

[0078] Overall, this series of technology integrations and checks significantly improve the quality and reliability of project reports. By systematically integrating fixed and variable contents, the reports are consistent in format and information, making information processing in a more complex business environment efficient. In addition, through consistency checking and accuracy evaluation, data errors can be detected and corrected in a timely manner, thus ensuring that decisions are based on the most reliable information. At the same time, the optimization of the template generation time not only speeds up the report production process but also enhances the team's response ability, enabling it to cope with challenges in a rapidly changing market. This comprehensive technical means improves work efficiency while endowing the team with stronger decision-making support capabilities, effectively driving the successful development of the enterprise.

[0079] Although embodiments of the present invention have been shown and described, it will be understood by those of ordinary skill in the art that various changes, modifications, substitutions and variations can be made to these embodiments without departing from the principles and spirit of the present invention, and the scope of the present invention is defined by the appended claims and their equivalents.

Claims

1. An automated generation method for project due diligence reports based on multi-source data integration, characterized in that, It includes the following steps: S1. Preset a report template. For multiple reports, define the common content as fixed content, the specific content as variables, and number the variables; S2. After obtaining the required data from each data source in different ways, remove duplicate data and number the data, including obtaining enterprise information from the big data platform through an interface, obtaining mortgage information from the China Bond Depository and Clearing Corporation Limited through RPA, and obtaining project information by OCR to identify the content of contract documents; S3. Match the variable numbers defined in S1 with the numbers of the data obtained in S2 to form a mapping relationship; S4. Integrate the fixed content and variable content according to the preset template to form a feasibility study report in a unified format, and conduct a report consistency check to evaluate the generation time of the preset template and the accuracy rate of the feasibility study report.

2. The automated generation method of the project due diligence report based on multi-source data integration according to claim 1, characterized in that: The formula for the variable numbering is as follows: V n = V base + n In the formula, V n represents the number of the n-th variable, and V base represents the initial value of the variable number, and n represents the serial number of the variable.

3. The automated generation method of the project due diligence report based on multi-source data integration according to claim 2, wherein: The formula for the data acquisition is as follows: D i = f(S i,j , T k ) In the formula, D i represents the data obtained from the i-th data source, and S i,j represents the j-th type of information in the i-th data source, and T k represents the data acquisition method.

4. The automated generation method of the project due diligence report based on multi-source data integration according to claim 3, characterized in that: The formula for removing duplicate data is as follows: D unique = D - D dup In the formula, D unique represents the valid data retained after the deduplication operation, D represents all the originally obtained data items, and D dup represents the set of duplicate data.

5. A method for automatically generating a project due diligence report based on multi-source data integration according to claim 4, characterized in that: The formula for numbering the data is as follows: N k = D source (k) In the formula, N k represents the number of the k-th acquired data, and D source (k) represents the k-th information extracted from the data source.

6. The automated generation method of the project due diligence report based on multi-source data integration according to claim 5, characterized in that: The mapping relationship is as follows: M(V n ,N k )→(matched) In the formula, M represents the mapping relation function, and V n represents the variable number defined in the report template, N represents the corresponding data number obtained from the data source, and matched represents the returned matching result.

7. A method for automatically generating a project due diligence report based on multi-source data integration according to claim 6, characterized in that: The formula for integrating the fixed content and variable content is as follows: R = C f + C v In the formula, R represents the finally generated report result, and C f represents the fixed content in the report template, and C v represents the variable content filled according to the mapping relationship.

8. A method for automatically generating a project due diligence report based on multi-source data integration according to claim 7, characterized in that: The formula for the report consistency check is as follows: Check(R) = {R min , R max , R avg} In the formula, Check(R) represents the consistency check function, where R min represents the minimum value of the data in the report, and R max represents the maximum value of the data in the report, and R avg represents the average value of the data in the report.

9. The automated generation method of the project due diligence report based on multi-source data integration according to claim 8, wherein: The formula for evaluating the generation time of the preset template is as follows: In the formula, T g represents the total time required to generate the template, T j represents the time required to generate each partial template, n represents the different parts combined in the report, and j represents the index subscript.

10. A method for automatically generating a project due diligence report based on multi-source data integration according to claim 9, characterized in that: The formula for evaluating the accuracy rate of the feasibility study report is as follows: In the formula, Acc represents the accuracy rate of the feasibility study report, and C correct represents the number of correctly matched contents, and C total represents the number of all contents in the report, including both the matched and unmatched contents.

Images