Method and system for performing intelligent form review on project application material
By using OCR and NLP technologies for data preprocessing and combining them with machine learning models for intelligent review, the problems of low efficiency, poor accuracy, and high labor costs in reviewing project application materials have been solved, achieving an efficient and accurate intelligent review process.
Patent Information
- Application Number
- CN202511485003.3
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-17
- Publication Date
- 2026-01-30
AI Technical Summary
Existing technologies suffer from problems such as low efficiency in reviewing project application materials, high accuracy due to subjective factors, high labor costs, and a lack of intelligent auxiliary tools.
Data collection and preprocessing are performed using Optical Character Recognition (OCR) technology and Natural Language Processing (NLP) technology. A review rule base is established, and intelligent review is performed using a machine learning text classification model. Review reports are automatically generated and the results are fed back.
This will significantly improve review efficiency, reduce subjective interference, enhance accuracy, lower labor costs, and ensure the fairness and rationality of resource allocation.
Smart Images
Figure CN121436902A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of project management technology, specifically providing a method and system for intelligent formal review of project application materials. Background Technology
[0002] In the current review of application materials for science and technology projects, manual review is the primary method. Reviewers need to examine each application material individually based on various standards, including project application conditions, application categories, material requirements, and veto criteria. The workflow is roughly as follows: First, the application materials are received, typically including the project application form, relevant supporting documents, and other documents. Then, reviewers manually review these materials to verify the applicant's legal status, such as whether their business license and research qualifications are complete and meet requirements; check the completeness of the application materials, such as whether key research plans and budget details are missing; and examine the logical relationships within the materials, such as the reasonable correlation between the expected project outcomes and the research content, and whether there are any contradictions between system records and the information submitted before and after the materials were submitted.
[0003] The problems existing in the prior art are as follows:
[0004] (1) Low review efficiency: Manual review requires reviewers to go through a large number of application materials page by page. For complex application projects, reviewing a single document may take several hours or even longer. Faced with a large number of application projects, the review cycle will be greatly extended, making it impossible to provide timely feedback to the applicants and seriously affecting the progress of science and technology projects.
[0005] (2) Accuracy is greatly affected by subjective factors: Different reviewers have different understandings and grasps of the review standards, and are easily influenced by personal subjective factors during the review process, resulting in inconsistent and inaccurate review results. This may cause some eligible projects to be misjudged due to review errors, while some ineligible projects may pass the review, undermining the fairness and rationality of the allocation of science and technology project resources.
[0006] (3) High labor costs: In order to complete the review of a large number of application materials within the stipulated time, the science and technology management department needs to invest a lot of manpower and organize professional reviewers to conduct centralized reviews. This not only requires paying high labor costs, but may also affect the review progress and quality due to insufficient manpower.
[0007] Therefore, how to solve the problems of low review efficiency, high accuracy due to subjective factors, high labor costs, and lack of intelligent auxiliary tools in the existing technology is an urgent problem to be solved by those skilled in the art. Summary of the Invention
[0008] This invention addresses the shortcomings of the prior art by providing a highly practical method for intelligent formal review of project application materials.
[0009] A further technical objective of this invention is to provide a reasonably designed, safe, and applicable device for intelligent formal review of project application materials.
[0010] The technical solution adopted by this invention to solve its technical problem is:
[0011] A method for intelligent formal review of project application materials includes the following steps:
[0012] S1. Data Acquisition and Preprocessing;
[0013] S2, Review rule configuration;
[0014] S3, Intelligent Review Execution;
[0015] S4. Generation and feedback of review results.
[0016] Furthermore, in step S1, optical character recognition (OCR) technology and natural language processing (NLP) technology are used to digitize the application materials and extract key information.
[0017] Furthermore, in step S2, a review rule base is established, and corresponding review rules are set for different types of science and technology projects, application conditions, and review points. Different review permissions are assigned to reviewers, and reviewers modify and improve the review rules in a timely manner according to changes in science and technology projects and adjustments to application requirements.
[0018] Furthermore, in step S3, the preprocessed application material information is compared and analyzed with the review rule base. A text classification model based on machine learning is used to determine whether the application materials comply with various review rules.
[0019] Furthermore, in step S4, a detailed review report is automatically generated based on the review results. The report includes whether the application materials have passed the review, and if not, it clearly points out the specific problems and the corresponding review rules and clauses.
[0020] Meanwhile, the review results will be fed back in a timely manner through the system.
[0021] A system for intelligent formal review of project application materials first collects and preprocesses data, then configures review rules, performs intelligent review, and finally generates and provides feedback on the review results.
[0022] Furthermore, during data collection and preprocessing, Optical Character Recognition (OCR) technology and Natural Language Processing (NLP) technology are used to digitize the application materials and extract key information.
[0023] Furthermore, when configuring review rules, a review rule library is established. Corresponding review rules are set for different types of science and technology projects, application conditions, and review points. Different review permissions are assigned to reviewers, and reviewers can promptly modify and improve the review rules according to changes in science and technology projects and adjustments to application requirements.
[0024] Furthermore, during the intelligent review process, the pre-processed application materials are compared and analyzed with the review rule base. A text classification model based on machine learning is used to determine whether the application materials comply with various review rules.
[0025] Furthermore, when generating and providing feedback on the review results, a detailed review report is automatically generated based on the review execution results. The report includes whether the application materials have passed the review, and if not, it clearly points out the specific problems and the corresponding review rules and clauses.
[0026] Meanwhile, the review results will be fed back in a timely manner through the system.
[0027] Compared with existing technologies, the intelligent formal review method and system for project application materials of the present invention have the following outstanding advantages:
[0028] (1) This invention significantly shortens the review time for application materials through an automated intelligent review process. Compared with traditional manual review, the review efficiency can be increased by several times or even dozens of times.
[0029] (2) Reviewing based on a unified review rule base and intelligent algorithm model avoids the interference of subjective factors in manual review, and improves the consistency and accuracy of review results. It effectively reduces the occurrence of misjudgments and ensures the fairness and rationality of the allocation of science and technology project resources.
[0030] (3) This invention reduces the reliance on a large number of manual reviewers. The science and technology management department only needs a small number of professionals to review and manage the review results, thus reducing the expenditure on human resources.
[0031] (4) It realizes intelligent analysis and review of application materials, which can quickly identify potential problems and risks in the materials and provide decision support for reviewers. Attached Figure Description
[0032] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0033] Figure 1 This is a flowchart illustrating a method for intelligent formal review of project application materials;
[0034] Figure 2 This is a schematic diagram of a framework for an intelligent formal review method for project application materials. Detailed Implementation
[0035] To enable those skilled in the art to better understand the present invention, the present invention will be further described in detail below with reference to specific embodiments. Obviously, the described embodiments are merely some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.
[0036] The following is a preferred embodiment:
[0037] like Figure 1 , 2 As shown in this embodiment, a method for intelligent formal review of project application materials includes the following steps:
[0038] S1. Data Acquisition and Preprocessing;
[0039] Optical Character Recognition (OCR) and Natural Language Processing (NLP) technologies are used to digitize application materials and extract key information. For example, information such as project name, applicant organization, principal investigator, research content, and budget details are extracted from the PDF document of the application. The extracted information is then cleaned and standardized, unifying the format of dates, amounts, and other data from different formats, and correcting typos and other errors.
[0040] S2, Review rule configuration;
[0041] Establish a flexibly configurable review rule library, and set corresponding review rules for different types of science and technology program projects, application conditions, and review points.
[0042] For example, the rules for reviewing the eligibility of applicant entities set specific review indicators such as the validity period of the business license and the relevance of the business scope to the project; the rules for reviewing the completeness of application materials clearly stipulate the types of documents that must be included and the key chapters. Reviewers can adjust and improve the rules according to actual needs.
[0043] S3, Intelligent Review Execution;
[0044] The pre-processed application materials are compared and analyzed against the review rule base. A text classification model based on machine learning is used to determine whether the application materials comply with various review rules.
[0045] For example, the Support Vector Machine (SVM) model can be used to classify the research content in the application materials to determine whether it belongs to the field specified in the project application category; and a rule engine can be established to conduct a compliance review of the budget details and check whether the expenditure of each expense meets the budget preparation requirements.
[0046] S4. Review results generation and feedback;
[0047] Based on the review results, a detailed review report is automatically generated. The report includes whether the application materials passed the review; if not, it clearly identifies the specific problems and the corresponding review rules and clauses. Simultaneously, the review results are promptly fed back through the system.
[0048] For example, the review report pointed out that the cost of a certain equipment purchase in the budget details of the application exceeded the prescribed standard, and the applicant needed to recalculate and adjust it.
[0049] Meanwhile, the module provides reviewers with functions to view, edit (and manually adjust when necessary), and export review results, facilitating subsequent processing of the results. Furthermore, this module is responsible for providing feedback on the review results to the applicant organization.
[0050] Different review permissions are assigned to reviewers. For example, ordinary reviewers can only review materials, while review supervisors have the authority to review review results and configure rules. Registration and login functions are provided for applicants to facilitate the submission of application materials and the query of review results.
[0051] It supports uploading application materials in various formats, such as PDF and WORD, and encrypts and stores the uploaded materials to ensure data security. It also provides version management functionality for application materials, allowing applicants to modify and update their materials during the review process.
[0052] Reviewers can promptly modify and improve the review rules based on policy changes and adjustments to application requirements for science and technology programs. For example, when the intellectual property requirements for a certain type of science and technology program change, reviewers can adjust the relevant intellectual property review rules accordingly in this module.
[0053] Based on the above method, the system for intelligent formal review of project application materials in this embodiment firstly collects and preprocesses data, then configures review rules, performs intelligent review, and finally generates and provides feedback on review results.
[0054] During data collection and preprocessing, optical character recognition (OCR) technology and natural language processing (NLP) technology are used to digitize the application materials and extract key information.
[0055] When configuring review rules, a review rule library is established. Corresponding review rules are set for different types of science and technology projects, application conditions, and review points. Different review permissions are assigned to reviewers. Reviewers modify and improve the review rules in a timely manner according to changes in science and technology projects and adjustments to application requirements.
[0056] During intelligent review, the pre-processed application materials are compared and analyzed with the review rule base. A text classification model based on machine learning is used to determine whether the application materials comply with various review rules.
[0057] When generating and providing feedback on review results, a detailed review report is automatically generated based on the review results. The report includes whether the application materials have passed the review, and if not, it clearly points out the specific problems and the corresponding review rules and clauses.
[0058] Meanwhile, the review results will be fed back in a timely manner through the system.
[0059] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.
Claims
1. A method for intelligent formal review of project application materials, characterized in that, Having the following steps: S1, data collection and pretreatment; S2, review rule configuration; S3, intelligent review execution; S4, review result generation and feedback.
2. The method for intelligent form review of project declaration materials according to claim 1, characterized in that, In step S1, the optical character recognition (OCR) technology and natural language processing (NLP) technology are used to digitize the declaration materials and extract key information.
3. The method for intelligent form review of project declaration materials according to claim 2, characterized in that, In step S2, a review rule library is established, and corresponding review rules are set for different types of science and technology plan projects, declaration conditions and review points. Different review permissions are assigned to review personnel, and review personnel can modify and improve the review rules in a timely manner according to the changes of science and technology plan projects and the adjustment of declaration requirements.
4. The method for intelligent form review of project declaration materials according to claim 3, characterized in that, In step S3, the pretreated declaration material information is compared and analyzed with the review rule library, and a text classification model based on machine learning is used to judge whether the declaration material meets the review rules.
5. The method for intelligent form review of project declaration materials according to claim 4, characterized in that, In step S4, a detailed review report is automatically generated according to the review execution result, including whether the declaration material passes the review, and if not, the specific problems and corresponding review rule clauses are specified. At the same time, the review result is fed back through the system in a timely manner.
6. A system for intelligent form review of project submissions, the system comprising: First, data collection and pretreatment are performed, then review rule configuration, intelligent review execution, and finally review result generation and feedback.
7. The system for intelligent form review of project declaration materials of claim 6, wherein, During data collection and pretreatment, optical character recognition (OCR) technology and natural language processing (NLP) technology are used to digitize the declaration materials and extract key information.
8. The system for intelligent form review of project declaration materials of claim 7, wherein, During review rule configuration, a review rule library is established, and corresponding review rules are set for different types of science and technology plan projects, declaration conditions and review points. Different review permissions are assigned to review personnel, and review personnel can modify and improve the review rules in a timely manner according to the changes of science and technology plan projects and the adjustment of declaration requirements.
9. The system for intelligent form review of project declaration materials of claim 8, wherein, During intelligent review execution, the pretreated declaration material information is compared and analyzed with the review rule library, and a text classification model based on machine learning is used to judge whether the declaration material meets the review rules.
10. The system for intelligent form review of project declaration materials of claim 8, wherein, During review result generation and feedback, a detailed review report is automatically generated according to the review execution result, including whether the declaration material passes the review, and if not, the specific problems and corresponding review rule clauses are specified. At the same time, the review result is fed back through the system in a timely manner.