Blockchain-based scientific and technological achievements authenticity verification and evaluation platform

Abstract

The present invention discloses a blockchain-based scientific and technological achievement authenticity verification and evaluation platform, which relates to the technical field of scientific and technological achievement management. The present invention comprises an upload module, a preliminary screening module, a verification module, an allocation module and a comprehensive evaluation module; the upload module is used to provide an interface and a variety of upload methods, optimize the upload interface to ensure stable and efficient upload of files of different sizes, and use a hash algorithm to compare the file hash value with a pre-stored standard value to judge the file integrity, thereby ensuring the accuracy and completeness of the uploaded file; the present invention ensures the authenticity and reliability of data through a variety of technical means, the upload module uses a hash algorithm to ensure the file integrity, the preliminary screening module uses natural language and image recognition technology to assist in judging the validity of materials, the verification module traces back and cross-verifies the data source based on the blockchain, and the synergistic effect of multiple links effectively avoids data problems, provides a solid foundation for subsequent evaluation, and enhances the credibility and accuracy of data.

Description

Technical Field

[0001] The present invention relates to the technical field of scientific and technological achievement management, and specifically to a blockchain-based scientific and technological achievement authenticity verification and evaluation platform. Background Art

[0002] In today's era of rapid technological development, the number of scientific and technological achievements is increasing day by day, and their management is becoming more and more important and facing many challenges. The management of scientific and technological achievements covers multiple links from the generation, application, verification, evaluation to transformation and application of achievements, among which authenticity verification and objective evaluation are the key to ensuring the quality and value of scientific and technological achievements. However, the traditional management model of scientific and technological achievements has exposed many defects in these two aspects.

[0003] In terms of authenticity verification, the reliability of data sources is difficult to guarantee. To maintain competitive advantage or for other reasons, the owners of the results may submit incomplete or inaccurate data, and some data may even be falsified. Existing management methods make it difficult to effectively trace and verify the source of data. At the same time, the lack of unified and strict data verification standards and processes makes it difficult to compare the authenticity of different results horizontally.

[0004] In the evaluation process, the problems are also significant. The evaluation standards are not unified, and the evaluation indicators adopted by different regions, institutions and industries vary greatly, which means that the same scientific and technological achievements may produce completely different results under different evaluation systems, lacking comparability. In the evaluation process, the expert selection mechanism is imperfect, and there are interferences from human factors, which may lead to experts not having sufficient professional capabilities or objectivity. Moreover, the evaluation process has low transparency, and data is difficult to share and trace, which can easily lead to a crisis of trust and affect the transformation and promotion of scientific and technological achievements.

[0005] Although there are some solutions to the above problems, there are still obvious shortcomings. For example, some achievement management platforms try to establish databases to store achievement information, but the security and integrity of the data cannot be effectively guaranteed, and the risk of data leakage and tampering is high. Some institutions rely on manual review and paper materials during the evaluation process, which is not only inefficient, but also prone to subjective bias and human errors.

[0006] In addition, in terms of data verification, existing technical means are mostly limited to simple data format checks and surface information verification, making it difficult to deeply explore the authenticity and reliability behind the data. For cross-domain and highly comprehensive scientific and technological achievements, existing evaluation methods are often unable to comprehensively and accurately measure their value because these methods lack the ability to comprehensively consider and dynamically adjust multi-dimensional indicators.

[0007] In order to solve the above defects, a technical solution is now provided. Summary of the invention

[0008] The purpose of this invention is to solve the problems of difficulty in authenticity verification, inconsistent evaluation standards, unscientific expert selection, lack of process transparency and inability to guarantee data security in traditional scientific and technological achievement management, and to propose a blockchain-based scientific and technological achievement authenticity verification and evaluation platform.

[0009] The purpose of the present invention can be achieved through the following technical solutions:

[0010] The blockchain-based scientific and technological achievements authenticity verification and evaluation platform includes:

[0011] The upload module is used to provide an interface and multiple upload methods, optimize the upload interface to ensure stable and efficient upload of files of different sizes, and use the hash algorithm to compare the file hash value with the pre-stored standard value to determine the file integrity;

[0012] The initial screening module screens scientific and technological achievement materials based on the rules jointly formulated by administrators and experts, including keyword matching and file format restrictions. At the same time, it uses natural language processing and image recognition technology to extract key information to assist in judging the validity, and preliminarily excludes invalid or non-compliant materials;

[0013] The verification module uses blockchain to trace the source of data, cross-verifies the reliability with authoritative data sources, determines the weights through hierarchical analysis, and calculates the comprehensive authenticity value by weighted summation to judge the authenticity of scientific and technological achievement materials;

[0014] The allocation module selects experts suitable for reviewing different results based on the information in the expert database, and prioritizes the tasks by calculating the remaining workload based on their current workload and estimated completion time. At the same time, it combines professional matching and historical evaluation efficiency, and estimates the submission time based on the complexity of the results, the workload of the experts, and their capabilities during the review, and regularly reminds the experts;

[0015] The comprehensive evaluation module establishes an evaluation system with multiple indicators and dynamically adjusted weights. The comprehensive evaluation score is calculated by integrating the comprehensive authenticity value of the verification module with the weighted average of the expert indicator scores. The consistency of expert opinions is measured by the Kendall harmony coefficient. Finally, a detailed evaluation report is generated to comprehensively and objectively evaluate scientific and technological achievements.

[0016] Furthermore, the execution steps of the upload module are as follows:

[0017] Provides a user upload interface that supports a variety of common file upload methods, including drag-and-drop upload and click-to-upload. At the same time, in order to adapt to files of different sizes, the upload is optimized to ensure the stability and efficiency of the upload process.

[0018] Use the hash algorithm to calculate the uploaded file and generate the hash value of the file;

[0019] At the same time, the standard hash value of each file is pre-stored in the upload module, and the integrity of the file is judged by comparing the hash value of the uploaded file with the standard hash value; when the hash value does not match, the user is prompted to upload again.

[0020] Furthermore, the specific operation steps of the upload module to optimize the upload are as follows:

[0021] Optimize through asynchronous upload mechanism and block upload strategy. Asynchronous upload method is adopted to make file upload operation in background thread, avoid blocking the main thread and ensure the responsiveness of user interface. Users can perform other operations synchronously while uploading files, which improves user experience.

[0022] For large files, split them into multiple data blocks for uploading. The size of each data block is set to a fixed value. For large files, if the file memory exceeds the preset threshold, it is considered a large file.

[0023] Through the block upload strategy, if a data block fails to be uploaded, only the data block needs to be uploaded again instead of the entire file, saving time and network resources. Only data blocks are processed each time, which reduces the peak memory usage and avoids memory overflow caused by uploading large files.

[0024] On the basis of block uploading, the breakpoint resume function is implemented to record the information of the successfully uploaded data blocks. When the upload process is interrupted and then resumed, the remaining data blocks are uploaded from the last interrupted position. Specifically: the progress information of the file upload is recorded on the server and client respectively, including the uploaded data block number and the hash value of each data block. When the user re-initiates the upload request, the client sends a resume request to the server, carrying the uploaded data block information. The server determines which data blocks need to be re-uploaded based on this information, thereby resuming the upload from the breakpoint.

[0025] Furthermore, the execution process of the initial screening module is as follows:

[0026] Establish a rule-based screening mechanism. The rules are jointly formulated by the platform administrator and domain experts. The rules include that scientific and technological achievement materials must contain corresponding keywords. If these keywords are missing, they are judged invalid. Specifically, the document is represented as a vector space model, and the similarity between the document vector and the keyword vector is calculated. Let the keyword set contained in document D be K D ={k 1 , k 2 , …, k n}, the weight of each keyword is The keyword vector is represented as

[0027] For the corresponding keyword set KT ={k t1 , k t2 , …, k tm}, whose vector is represented as

[0028] Then the similarity between the document and the corresponding keyword set is Sim(D, K T ) is calculated by the cosine similarity formula: Set a similarity threshold θ, when Sim(D, K T )≥θ, the document is considered to contain enough corresponding keywords and pass the initial screening; otherwise, it is considered invalid;

[0029] For file formats, only files of preset types are allowed to pass the initial screening;

[0030] Use natural language processing technology and image recognition technology to extract key information from documents to further assist in determining whether the materials are valid; use natural language processing technology to analyze document content to determine whether clear scientific and technological achievements content and logic are described.

[0031] Furthermore, the specific operation steps of the initial screening module to extract key information from the file and analyze the document content are as follows:

[0032] Natural language processing: Convert all document formats into plain text, remove irrelevant content and characters in headers and footers, and extra spaces and line breaks; use a word segmentation algorithm to segment text and mark parts of speech, which is specifically achieved through the NLTK open source tool; remove meaningless words based on the stop word list to reduce the amount of data;

[0033] Formulate rules based on the characteristics of scientific and technological achievements to extract keywords including professional terms and noun phrases; use LDA topic model to determine document topics and related keywords; identify scientific research institutions, technology names, and time entities as key information;

[0034] Develop templates based on common logical structures, match documents to determine logical integrity, use semantic analysis technology to clarify the semantic relationship between words and sentences, and determine the logical rationality of the results; and analyze sentence coherence to determine the fluency and logic of the document's narrative;

[0035] Image recognition technology: unify image formats, adjust resolution and size; use deep learning algorithms to identify target objects in images, extract image text and obtain key information through text processing; calculate image feature vectors and classify them, and extract key information by category;

[0036] Integrate key information extracted from images and text to form more complete results information, and use image information to assist in judging the validity of materials.

[0037] Furthermore, the specific operation steps of the verification module are as follows:

[0038] Using the traceability of blockchain, trace the source of data related to scientific and technological achievements; check whether the data comes from reliable channels, including whether it matches the data of known authoritative databases and research institutions; for experimental data, verify the rationality of its experimental design and the consistency of the data;

[0039] Data interaction and cross-validation are carried out with external authoritative data platforms and industry standard libraries. For the results of a new technology, its technical parameters are compared with the standard parameter range recognized in the industry. At the same time, comparative analysis is carried out in combination with the historical scientific and technological achievement data stored in the platform, and a comprehensive authenticity value is calculated through multi-dimensional verification. When the comprehensive authenticity value exceeds the preset threshold, the scientific and technological achievement material is judged to have passed the authenticity verification.

[0040] Furthermore, the authenticity comprehensive value judgment process in the verification module is as follows:

[0041] Use the hierarchical analysis method to determine the weight of each verification factor, and then sum the weighted values ​​to get the comprehensive value; suppose there are m verification factors, including the reliability of the data source, and the weight is w 1 , experimental data rationality, assuming the weight is w 2 , and the degree of compliance with industry standards, assuming the weight is w 3 , and their corresponding verification scores are s 1 ,s 2 ,s 3 ,…,s m ; First, the judgment matrix is ​​constructed through the hierarchical analysis method, and the characteristic vector is solved to obtain the weight of each factor to meet The calculation formula of the authenticity comprehensive value V is: V = w 1 ×s 1 +w 2 ×s 2 +w 3 ×s 3 +…+w m ×s m ; Set a threshold V T , when V ≥ V T When the scientific and technological achievement materials are verified, it is judged that they have passed the authenticity verification.

[0042] Furthermore, the allocation module performs the following steps:

[0043] Establish a database containing information on experts in many fields, and classify and index the expert information; the information in the database includes the basic information of the experts, their research fields, technical directions they are good at, and the projects they have recently participated in the evaluation; through the expert information, select the experts who are suitable for reviewing the corresponding scientific and technological achievements;

[0044] Calculate the remaining workload of each expert based on the number of current review tasks and the expected completion time; when allocating new scientific and technological achievement materials, give priority to experts with the least remaining workload;

[0045] At the same time, the professional matching degree of the experts is taken into account to ensure that the materials assigned to the experts are within their professional fields. The algorithm is dynamically adjusted according to the historical evaluation efficiency of the experts to achieve a more reasonable allocation. When calculating the remaining workload of the experts and allocating scientific and technological achievement materials, the current number of review tasks of expert i is set to n. i , the average time to complete a single task is expected to be t i , then the remaining workload of expert i is W i Expressed as: W i =n i ×t i ; For a new scientific and technological achievement material, select the expert with the smallest remaining workload for assignment, that is, select W i The smallest expert i such that: W j =min{W 1 , W 2 , …, W k}; where k is the total number of experts; at the same time, combined with the expert professional matching degree, set it to M ij , indicating the professional matching degree between experts and scientific and technological achievements. ij When the number is greater than the preset threshold, experts will be assigned to review the scientific and technological achievement materials;

[0046] During the review, a standard review time is given according to the complexity of the scientific and technological achievement materials. Combined with the current review workload and capabilities of the assigned experts, the submission time of the scientific and technological achievement materials after expert review is determined, and reminders are issued regularly to remind experts to review the scientific and technological achievement materials.

[0047] Furthermore, the process of determining the submission time of scientific and technological achievement materials after being reviewed by experts and issuing reminders regularly in the allocation module is as follows:

[0048] The complexity assessment process of scientific and technological achievement materials is as follows:

[0049] Multiple evaluation indicators are selected to measure the complexity of scientific and technological achievement materials, including:

[0050] Technology depth index GS: reflects the degree of sophistication and difficulty of the technology involved in the achievement. It is determined by analyzing the new technologies used in the achievement, the number of innovative points, and the cutting-edge factors of the technology, and is quantified by assigning a value of 1-100;

[0051] Document size index RE: The number of pages, words, and figures in the document are used as the measurement criteria, and a value of 0-1 is assigned to it;

[0052] Field breadth indicator QE: Analyze the number of disciplines and the degree of intersection involved in the results, use this as the basis for evaluation, and assign a score of 1-10;

[0053] Then set weights for each indicator and determine their relative importance based on experience and data analysis; calculate the complexity score of scientific and technological achievement materials based on the selected indicators and weights, using the following formula: 分 =GS×ψ 1 +RE×ψ 2 +QE×ψ 3 , where ψ 1 , 2 , 3 They are the weights set for the technical depth index, document size index, and domain breadth index respectively;

[0054] Collect historical scientific and technological achievements review data, analyze the relationship between different complexity scores and actual review time, establish a mapping function between complexity score and standard review time through data fitting method, and determine the standard review time of the scientific and technological achievements based on this function;

[0055] After receiving new scientific and technological achievement materials, experts will process tasks in order and the efficiency of processing new tasks is the same as the efficiency of processing current tasks. In this case, the submission time of the scientific and technological achievement materials after being reviewed by experts is the total time required for experts to complete all current tasks plus the standard review time of the scientific and technological achievements;

[0056] According to the urgency and review cycle of the scientific and technological achievement materials, the reminder time interval is set, and reminders are issued 25%, 50% and 75% before the expected submission time respectively; according to the calculated submission time and the set reminder time interval, reminder information is automatically generated, and the reminder information includes the name of the scientific and technological achievement materials, the expert's current review progress, the estimated remaining review time and the content of the next reminder time; and it is sent to experts regularly via email, platform station message or mobile phone push to remind them to complete the review task on time.

[0057] Furthermore, the process performed by the comprehensive evaluation module is as follows:

[0058] Establish a comprehensive evaluation index system and assign corresponding weights to each evaluation index; these indicators include the novelty of the technology, technical difficulty, degree of improvement over existing technologies, potential market size, and environmental impact; the index weights are dynamically adjusted based on industry characteristics and market demand factors;

[0059] The authenticity comprehensive value calculated by the verification module is combined with the scores of each indicator given by the expert review; the weighted average method is used to combine the two parts of the results according to the pre-set weights; specifically: the authenticity comprehensive value V of the verification module is combined with the scores of each indicator given by the expert review. Assume that there are p indicators in the comprehensive evaluation indicator system, including technical novelty, expert score e 1 , weight a 1 , technical difficulty, expert rating 2 , weight a 2 , potential market size, expert rating 3 , weight a 3 ; The calculation formula for the comprehensive evaluation score S is: S = a 1 ×c 1 +a 2 ×e 2 +a 3 ×e 3 +…+a p ×e p +b×V; where b is the weight of the authenticity comprehensive value in the comprehensive evaluation, and satisfies According to industry characteristics and market demand, we will determine a i and the value of b;

[0060] At the same time, the consistency of opinions of different experts is measured by using the Kendall harmony coefficient; suppose m experts score a certain indicator of v scientific and technological achievements, and the scoring matrix is ​​R = (r ij ), where r ij represents the score of expert i on achievement j;

[0061] First, calculate the score ranking of each achievement From 1 to m; then, calculate the Kendall harmony coefficient W: When W is less than the preset threshold, it is judged that the experts have large differences of opinion and the results need to be further analyzed and adjusted, including organizing expert discussions or re-evaluation;

[0062] Finally, a detailed evaluation report is generated based on the comprehensive evaluation results. The report includes evaluation scores, detailed evaluation of each indicator, analysis of strengths and weaknesses, and recommendations.

[0063] Compared with the prior art, the present invention has the following beneficial effects:

[0064] (1) The present invention uses a variety of technical means to ensure that the data is authentic and reliable. The upload module uses a hash algorithm to ensure file integrity. The initial screening module uses natural language and image recognition technology to assist in judging the validity of materials. The verification module traces and cross-verifies the data source based on blockchain. The synergy of multiple links effectively avoids data problems, provides a solid foundation for subsequent evaluation, and enhances data credibility and accuracy.

[0065] (2) In the present invention, the allocation module rationally allocates expert tasks based on multiple factors, estimates the review time based on the complexity of the results and prompts the experts to ensure efficient and orderly review. The comprehensive evaluation module scientifically integrates multi-source data to calculate scores, dynamically adjusts indicator weights and monitors the consistency of expert opinions, timely optimizes and handles differences, and improves the overall efficiency and quality of the evaluation process;

[0066] (3) The present invention uses blockchain technology to make the management process traceable and tamper-proof, enhance transparency, scientifically formulate screening and evaluation standards, and openly and transparently allocate experts and monitor opinions to avoid human interference; detailed evaluation reports provide objective basis for the application of results, overall guarantee the fairness of platform evaluation, and help all parties make accurate judgments and decisions. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] In order to facilitate understanding by those skilled in the art, the present invention is further described below in conjunction with the accompanying drawings;

[0068] Figure 1 This is the overall system block diagram of the present invention. DETAILED DESCRIPTION

[0069] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0070] It should be understood that the terms "include" and "comprising" used in the specification and claims of the present disclosure indicate the presence of described features, integers, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integers, steps, operations, elements, components and / or collections thereof.

[0071] It should also be understood that the terms used in this disclosure are only for the purpose of describing specific embodiments and are not intended to limit the disclosure. As used in this disclosure and claims, the singular forms of "a", "an", and "the" are intended to include the plural forms unless the context clearly indicates otherwise. It should also be further understood that the term "and / or" used in this disclosure and claims refers to any combination of one or more of the associated listed items and all possible combinations, including these combinations.

[0072] like Figure 1 As shown, the blockchain-based scientific and technological achievement authenticity verification and evaluation platform includes an upload module, a preliminary screening module, a verification module, an allocation module and a comprehensive evaluation module;

[0073] The upload module is used to provide an interface and multiple upload methods, optimize the upload interface to ensure stable and efficient upload of files of different sizes, and use the hash algorithm to compare the file hash value with the pre-stored standard value to determine the file integrity, ensuring the accuracy and completeness of the uploaded file;

[0074] Provides a user-friendly upload interface and supports a variety of common file upload methods, including drag-and-drop upload and click-to-upload. At the same time, in order to adapt to files of different sizes, the upload is optimized to ensure the stability and efficiency of the upload process.

[0075] Specifically, the optimization is carried out through asynchronous upload mechanism and block upload strategy. The asynchronous upload method is adopted to make the file upload operation be performed in the background thread to avoid blocking the main thread and ensure the responsiveness of the user interface. Users can still perform other operations while uploading files, such as browsing the platform and viewing historical records, to improve the user experience. For large files, block upload divides them into multiple smaller data blocks for upload, and sets the size of each data block to a fixed value (such as 1MB or 2MB). Large files are defined as files exceeding the preset threshold. The block upload strategy can be used to achieve that if a data block fails to be uploaded, only the data block needs to be uploaded again, without uploading the entire file again. Save time and network resources; only process small data blocks each time, which reduces the peak memory usage and avoids memory overflow caused by uploading large files, thereby improving system stability; based on block uploading, implement breakpoint resume function, record the successfully uploaded data block information, and when the upload process is interrupted (such as network failure, user pause, etc.) and resumed, continue uploading the remaining data blocks from the last interrupted position. Specifically, the progress information of the file upload is recorded on the server and client, including the uploaded data block number, the hash value of each data block, etc. When the user re-initiates the upload request, the client sends a resume request to the server, carrying the uploaded data block information. The server determines which data blocks need to be re-uploaded and which can be skipped directly based on this information, so as to resume uploading from the breakpoint;

[0076] A hash algorithm is used to calculate the uploaded file to generate the hash value of the file; at the same time, the standard hash value of each file is pre-stored in the upload module (generated according to the file template when the system is initialized), and the integrity of the file is judged by comparing the hash value of the uploaded file with the standard hash value; if the hash value does not match, the user is prompted to upload again.

[0077] The initial screening module screens scientific and technological achievement materials based on the rules jointly formulated by administrators and experts, including keyword matching and file format restrictions. At the same time, it uses natural language processing and image recognition technology to extract key information to assist in judging the validity, and preliminarily excludes invalid or non-compliant materials;

[0078] Establish a rule-based screening mechanism. These rules are jointly formulated by platform administrators and domain experts. It is stipulated that scientific and technological achievement materials must contain specific keywords (such as technical innovation points, application scenarios, etc.). If these keywords are missing, they are judged as invalid. Specifically, the document is represented as a vector space model, and the similarity between the document vector and the keyword vector is calculated. Let the keyword set contained in document D be K D ={k 1 , k 2 , …, k n}, the weight of each keyword is (calculated based on word frequency and inverse document frequency), the keyword vector is represented as For a specific keyword set K T ={k t1 , k t2 , …, k tm}, whose vector is represented as Then the similarity between the document and a specific keyword set is Sim(D, K T ) can be calculated using the cosine similarity formula: Set a similarity threshold θ. When Sim(D, KT)≥θ, the document is considered to contain enough specific keywords and pass the initial screening; otherwise, it is considered invalid. For file formats, only specific types of files are allowed to pass the initial screening, such as documents in .docx, .pdf and other formats.

[0079] Use natural language processing technology and image recognition technology (for materials containing images) to extract key information from the document to further assist in determining whether the material is valid; use natural language processing technology to analyze the document content to determine whether it describes clear scientific and technological achievements content and logic; the specific process is as follows:

[0080] Natural language processing: Convert all document formats into plain text, remove irrelevant content such as headers and footers, special characters, and extra spaces and line breaks; use word segmentation algorithms to segment text and mark parts of speech, which can be done with the help of open source tools such as NLTK; remove meaningless words based on stop word lists to reduce the amount of data; formulate rules based on the characteristics of the field of scientific and technological achievements, and extract keywords such as professional terms and noun phrases; use topic models such as LDA to determine document topics and related keywords; identify entities such as scientific research institutions, technology names, and time as key information; formulate templates based on common logical structures, match documents to determine logical integrity, and use semantic analysis technology to clarify the semantic relationship between words and sentences and determine the logical rationality of the results; use algorithms to analyze sentence coherence and determine the fluency and logic of document narration;

[0081] Image recognition technology: unify image formats, adjust resolution and size, and use related technologies to improve image quality and reduce noise interference; use deep learning algorithms to identify target objects in images, such as experimental equipment, etc., extract image text and obtain key information using text processing methods; calculate image feature vectors and classify them, and extract key information by category; integrate key information extracted from images and text to form more complete results information, and use image information to assist in judging the effectiveness of materials, such as the relevance of equipment and technology, and the consistency of graphic and text information.

[0082] The verification module uses blockchain to trace the source of data, cross-verifies the reliability with authoritative data sources, determines the weights through hierarchical analysis, and calculates the comprehensive authenticity value by weighted summation to judge the authenticity of scientific and technological achievement materials;

[0083] Using the traceability of blockchain, trace the source of data related to scientific and technological achievements; check whether the data comes from reliable channels, such as whether it matches the data of known authoritative databases, research institutions, etc.; for experimental data, verify the rationality of its experimental design and the consistency of the data;

[0084] Data interaction and cross-validation are carried out with external authoritative data platforms and industry standard libraries. For the results of a new technology, its technical parameters are compared with the standard parameter range recognized in the industry. At the same time, a comparative analysis is carried out in combination with the historical scientific and technological achievements data stored in the platform, and a comprehensive authenticity value is calculated through multi-dimensional verification. When this value exceeds the preset threshold, it indicates that the scientific and technological achievement material has passed the authenticity verification. Specifically, the analytic hierarchy process (AHP) is used to determine the weight of each verification factor, and then the weighted sum is obtained to obtain the comprehensive value. Assume that there are m verification factors, including data source reliability (the weight is w 1 ), rationality of experimental data (assuming weight is w 2 ), and compliance with industry standards (assuming the weight is w 3 ), etc., and their corresponding verification scores are s 1 ,s 2 ,s 3 ,…,s m ; First, the judgment matrix is ​​constructed through the hierarchical analysis method, and the characteristic vector is solved to obtain the weight of each factor to meet The calculation formula of the authenticity comprehensive value V is: V = w 1 ×s 1 +w 2 ×s 2 +w 3 ×s 3 +…+w m ×s m ; Set a threshold V T , when V ≥ V T When the scientific and technological achievement materials are verified, it is judged that they have passed the authenticity verification.

[0085] The allocation module selects experts suitable for reviewing different results based on the information in the expert database, and prioritizes tasks by calculating the remaining workload based on their current workload and estimated completion time. It also considers professional matching and historical evaluation efficiency. During the review, it estimates the submission time based on the complexity of the results, the workload of the experts, and their capabilities, and regularly reminds the experts.

[0086] Establish a database containing information about experts in many fields, and classify and index the expert information. The information in the database includes the expert's basic information, research field, technical direction, and recent projects involved in evaluation; through this information, screen out experts who are suitable for reviewing specific scientific and technological achievements; calculate the remaining workload of each expert based on factors such as the current number of review tasks and the expected completion time; when allocating new scientific and technological achievement materials, give priority to experts with the least remaining workload; at the same time, consider the professional matching of experts to ensure that the materials allocated to experts are within the scope of their professional fields; the algorithm is dynamically adjusted according to the expert's historical evaluation efficiency (such as average review time, evaluation quality, etc.) to achieve a more reasonable allocation; specifically, when calculating the expert's remaining workload and allocating scientific and technological achievement materials, assume that the current number of review tasks of expert i is n i , the average time to complete a single task is expected to be t i , then the remaining workload of expert i is W i Expressed as: W i =n i ×t i ; For a new scientific and technological achievement material, select the expert with the smallest remaining workload for assignment, that is, select W i The smallest expert j such that: W j =min{W 1 , W 2 , …, W k}; k is the total number of experts; at the same time, combined with the expert professional matching degree (set as M ij , which indicates the degree of professional matching between experts and scientific and technological achievements, with a value range of 0 to 1). ij When the number is greater than the preset threshold, experts will be assigned to review the scientific and technological achievement materials;

[0087] During the review, a standard review time is given according to the complexity of the scientific and technological achievement materials. Combined with the current review workload and capabilities of the assigned experts, the submission time of the scientific and technological achievement materials after expert review is determined, and prompts are issued regularly to remind experts to review the scientific and technological achievement materials. The complexity evaluation process of the scientific and technological achievement materials is as follows: multiple evaluation indicators are selected to measure the complexity of the scientific and technological achievement materials, including:

[0088] Technical depth index GS: reflects the degree of sophistication and difficulty of the technology involved in the results. It is determined by analyzing the new technologies used in the results, the number of innovative points and the cutting-edge factors of the technology, and is quantified by assigning a value of 1-100; Document size index RE: uses the number of pages, words, and figures in the document as measurement criteria, and comprehensively assigns a value of 0-1 for measurement; Field breadth index QE: considers the number of subject areas and the degree of intersection involved in the results, and uses this as the basis for evaluation, assigning a score of 1-10;

[0089] Then set weights for each indicator and determine their relative importance based on experience and data analysis; calculate the complexity score of scientific and technological achievement materials based on the selected indicators and weights, using the following formula: 分 =GS×ψ 1 +RE×ψ 2 +QE×ψ 3 , where ψ 1 , 2 , 3 The weights of the technical depth index, document size index and field breadth index are set respectively; historical scientific and technological achievements review data are collected, the relationship between different complexity scores and actual review time is analyzed, and a mapping function between complexity score and standard review time is established through data fitting method, and the standard review time of the scientific and technological achievements is determined according to the function;

[0090] After receiving new scientific and technological achievement materials, experts will process the tasks in order, and the efficiency of processing new tasks is the same as that of processing current tasks. The submission time of the scientific and technological achievement materials after expert review is the total time required for experts to complete all current tasks plus the standard review time of the scientific and technological achievements; according to the urgency and review cycle of the scientific and technological achievement materials, set a reasonable reminder time interval, and specifically send reminders before the estimated submission time of 25%, 50% and 75% respectively; the system automatically generates reminder information based on the calculated submission time and the set reminder time interval, and the reminder information includes the name of the scientific and technological achievement materials, the current review progress of the experts, the estimated remaining review time, and the time of the next reminder. It is sent to experts regularly through emails, platform station letters or mobile phone push notifications to remind them to complete the review tasks on time.

[0091] The comprehensive evaluation module establishes an evaluation system with multiple indicators and dynamically adjusted weights. The comprehensive evaluation score is calculated by integrating the comprehensive value of the verification module's authenticity with the weighted average of the expert indicator scores. The consistency of expert opinions is measured by the Kendall harmony coefficient. When there are large differences, further analysis and adjustment are carried out. Finally, a detailed evaluation report is generated to comprehensively and objectively evaluate scientific and technological achievements.

[0092] Establish a set of comprehensive evaluation index system and assign corresponding weights to each evaluation index; these indicators include but are not limited to the novelty of technology, technical difficulty, degree of improvement on existing technology, potential market size, and impact on the environment; the index weights are dynamically adjusted according to factors such as industry characteristics and market demand to ensure the scientificity and objectivity of the evaluation results; the authenticity comprehensive value calculated by the verification module is integrated with the scores of each index given by the expert review; the weighted average method is used to integrate the two parts of the results according to the pre-set weights; specifically: the authenticity comprehensive value V of the verification module is integrated with the scores of each index given by the expert review. Assume that there are p indicators in the comprehensive evaluation index system, such as technical novelty (expert score e 1 , weight a 1 ), technical difficulty (expert score e 2 , weight a 2 ), potential market size (expert rating e 3 , weight a 3 ) etc.; then the calculation formula of the comprehensive evaluation score S is: S = a 1 ×c 1 +a 2 ×c 2 +a 3 ×e 3 +…+a p ×e p +b×V; where b is the weight of the authenticity comprehensive value in the comprehensive evaluation, and satisfies According to industry characteristics and market demand, we will determine a i and the value of b;

[0093] At the same time, the consistency of opinions of different experts is shared. If there are large differences in expert opinions, the results are further analyzed and adjusted; Kendall's W is used for measurement; suppose m experts score a certain indicator of v scientific and technological achievements, and the scoring matrix is ​​R = (r ij ), where r ij represents the score of expert i on achievement j; first, calculate the score ranking of each achievement (from 1 to m); then, calculate the Kendall harmony coefficient W: The value range of W is 0 to 1. The closer it is to 1, the more consistent the expert opinions are. When it is less than the preset threshold, it is considered that the expert opinions are quite different and further analysis and adjustment of the results are needed, such as organizing expert discussions or re-evaluation.

[0094] Finally, a detailed evaluation report is generated based on the comprehensive evaluation results. The report content includes evaluation scores, detailed evaluation of each indicator, analysis of strengths and weaknesses, and recommendations.

[0095] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A blockchain-based scientific and technological achievement authenticity verification and evaluation platform, characterized by: include: The upload module is used to provide an interface and multiple upload methods, optimize the upload interface to ensure stable and efficient upload of files of different sizes, and use the hash algorithm to compare the file hash value with the pre-stored standard value to determine the file integrity. The execution steps are as follows: Provides a user upload interface that supports a variety of common file upload methods, including drag-and-drop upload and click-to-upload. At the same time, in order to adapt to files of different sizes, the upload is optimized to ensure the stability and efficiency of the upload process. Use the hash algorithm to calculate the uploaded file and generate the hash value of the file; At the same time, the standard hash value of each file is pre-stored in the upload module, and the integrity of the file is determined by comparing the hash value of the uploaded file with the standard hash value; when the hash value does not match, the user is prompted to re-upload; The specific operation steps of the upload module to optimize the upload are as follows: Optimize through asynchronous upload mechanism and block upload strategy. Asynchronous upload method is adopted to make file upload operation in background thread, avoid blocking the main thread and ensure the responsiveness of user interface. Users can perform other operations synchronously while uploading files, which improves user experience. For large files, split them into multiple data blocks for uploading. The size of each data block is set to a fixed value. For large files, if the file memory exceeds the preset threshold, it is considered a large file. Through the block upload strategy, if a data block fails to be uploaded, only the data block needs to be re-uploaded instead of the entire file, saving time and network resources; Only data blocks are processed each time, which reduces the peak memory usage and avoids memory overflow caused by uploading large files. On the basis of block uploading, the breakpoint resume function is implemented to record the information of the successfully uploaded data blocks. When the upload process is interrupted and then resumed, the remaining data blocks are uploaded from the position where the last interruption occurred. Specifically: the progress information of the file upload is recorded on the server and client respectively, including the uploaded data block number and the hash value of each data block. When the user re-initiates an upload request, the client sends a resume request to the server, carrying the uploaded data block information. The server determines which data blocks need to be re-uploaded based on this information, thereby resuming the upload from the breakpoint. The initial screening module screens scientific and technological achievement materials according to the rules jointly formulated by administrators and experts, including keyword matching and file format restrictions. At the same time, it uses natural language processing and image recognition technology to extract key information to assist in judging the validity, and preliminarily excludes invalid or non-compliant materials. The execution process is as follows: Establish a rule-based screening mechanism. The rules are jointly formulated by the platform administrator and domain experts. The rules include that scientific and technological achievement materials must contain corresponding keywords. If these keywords are missing, they are judged invalid. Specifically, the document is represented as a vector space model, and the similarity between the document vector and the keyword vector is calculated. Let the keyword set contained in document D be K D ={k1, k2, ..., k n }, the weight of each keyword is The keyword vector is represented as For the corresponding keyword set K T ={k t1 , k t2 , …, k tm }, whose vector is represented as Then the similarity between the document and the corresponding keyword set is Sim(D, K T ) is calculated by the cosine similarity formula: Set a similarity threshold θ, when Sim(D, K T )≥θ, the document is considered to contain enough corresponding keywords and pass the initial screening; otherwise, it is considered invalid; For file formats, only files of preset types are allowed to pass the initial screening; Use natural language processing technology and image recognition technology to extract key information from documents and further assist in determining whether the materials are valid; use natural language processing technology to analyze document content and determine whether clear scientific and technological achievements content and logic are described; The specific operation steps of the initial screening module to extract key information from the file and analyze the document content are as follows: Natural language processing: Convert all document formats into plain text, remove irrelevant content and characters in headers and footers, and extra spaces and line breaks; use a word segmentation algorithm to segment text and mark parts of speech, which is specifically achieved through the NLTK open source tool; remove meaningless words based on the stop word list to reduce the amount of data; Formulate rules based on the characteristics of scientific and technological achievements to extract keywords including professional terms and noun phrases; use LDA topic model to determine document topics and related keywords; identify scientific research institutions, technology names, and time entities as key information; Develop templates based on common logical structures, match documents to determine logical integrity, use semantic analysis technology to clarify the semantic relationship between words and sentences, and determine the logical rationality of the results; and analyze sentence coherence to determine the fluency and logic of the document's narrative; Image recognition technology: unify image formats, adjust resolution and size; Use deep learning algorithms to identify target objects in images, extract text from images and obtain key information using text processing methods; calculate image feature vectors and classify them, extracting key information by category; Integrate key information extracted from images and text to form more complete results information, and assist in judging the effectiveness of materials based on image information; The verification module uses blockchain to trace the source of data, cross-verifies the reliability with authoritative data sources, determines the weights through hierarchical analysis, and calculates the comprehensive authenticity value by weighted summation to judge the authenticity of scientific and technological achievement materials; The allocation module selects experts suitable for reviewing different results based on the information in the expert database, and prioritizes the tasks by calculating the remaining workload based on their current workload and estimated completion time. At the same time, it combines professional matching and historical evaluation efficiency, and estimates the submission time based on the complexity of the results, the workload of the experts, and their capabilities during the review, and regularly reminds the experts; The comprehensive evaluation module establishes an evaluation system with multiple indicators and dynamically adjusted weights. The comprehensive evaluation score is calculated by integrating the comprehensive authenticity value of the verification module with the weighted average of the expert indicator scores. The consistency of expert opinions is measured by the Kendall harmony coefficient. Finally, a detailed evaluation report is generated to comprehensively and objectively evaluate scientific and technological achievements.

2. The blockchain-based scientific and technological achievement authenticity verification and evaluation platform according to claim 1 is characterized in that: The specific operation steps of the verification module are as follows: Using the traceability of blockchain, trace the source of data related to scientific and technological achievements; check whether the data comes from reliable channels, including whether it matches the data of known authoritative databases and research institutions; for experimental data, verify the rationality of its experimental design and the consistency of the data; Data interaction and cross-validation are carried out with external authoritative data platforms and industry standard libraries. For the results of a new technology, its technical parameters are compared with the standard parameter range recognized in the industry. At the same time, the historical scientific and technological achievements data stored in the platform are combined for comparative analysis, and a comprehensive authenticity value is calculated through multi-dimensional verification. When the comprehensive authenticity value exceeds the preset threshold, the scientific and technological achievement material is judged to have passed the authenticity verification.

3. The blockchain-based scientific and technological achievement authenticity verification and evaluation platform according to claim 2 is characterized in that: The authenticity comprehensive value judgment process in the verification module is as follows: The analytic hierarchy process is used to determine the weight of each verification factor, and then the weighted sum is used to obtain the comprehensive value; assuming that there are m verification factors, including data source reliability, weight w1, experimental data rationality, weight w2, and compliance with industry standards, weight w3, the corresponding verification scores are s1, s2, s3, ..., s m ; First, the judgment matrix is ​​constructed through the hierarchical analysis method, and the characteristic vector is solved to obtain the weight of each factor to meet The calculation formula of the authenticity comprehensive value V is: V = w1×s1+w2×s2+w3×s3+…+w m ×s m ; Set a threshold V T , when V ≥ V T When the scientific and technological achievement materials are verified, it is judged that they have passed the authenticity verification.

4. The blockchain-based scientific and technological achievement authenticity verification and evaluation platform according to claim 1 is characterized in that: The allocation module execution process is as follows: Establish a database containing information on experts in many fields, and classify and index the expert information; the information in the database includes the basic information of the experts, their research fields, technical directions they are good at, and the projects they have recently participated in the evaluation; through the expert information, screen out experts who are suitable for reviewing the corresponding scientific and technological achievements; Calculate the remaining workload of each expert based on the number of current review tasks and the expected completion time; when allocating new scientific and technological achievement materials, give priority to experts with the least remaining workload; At the same time, the professional matching degree of the experts is taken into account to ensure that the materials assigned to the experts are within their professional fields. The algorithm is dynamically adjusted according to the historical evaluation efficiency of the experts to achieve a more reasonable allocation. When calculating the remaining workload of the experts and allocating scientific and technological achievement materials, the current number of review tasks of expert i is set to n. i , the average time to complete a single task is expected to be t i , then the remaining workload of expert i is W i Expressed as: W i =n i ×t i ; For a new scientific and technological achievement material, select the expert with the smallest remaining workload for assignment, that is, select W i The smallest expert j such that: W j =min{W1, W2, ..., W k }; where k is the total number of experts; at the same time, combined with the expert professional matching degree, set it to M ij , indicating the professional matching degree between experts and scientific and technological achievements. ij When the number of scientific and technological achievements exceeds the preset threshold, experts will be assigned to review the scientific and technological achievements materials; During the review, a standard review time is given according to the complexity of the scientific and technological achievement materials. Combined with the current review workload and capabilities of the assigned experts, the submission time of the scientific and technological achievement materials after expert review is determined, and reminders are issued regularly to remind experts to review the scientific and technological achievement materials.

5. The blockchain-based scientific and technological achievement authenticity verification and evaluation platform according to claim 4 is characterized in that: The process of determining the submission time of scientific and technological achievement materials after being reviewed by experts and issuing reminders regularly in the allocation module is as follows: The complexity assessment process of scientific and technological achievement materials is as follows: Multiple evaluation indicators are selected to measure the complexity of scientific and technological achievement materials, including: Technology depth index GS: reflects the degree of sophistication and difficulty of the technology involved in the achievement. It is determined by analyzing the new technologies used in the achievement, the number of innovations, and the cutting-edge factors of the technology, and is quantified by assigning a value of 1-100; Document size index RE: The number of pages, words, and figures in the document are used as the measurement criteria, and a value of 0-1 is assigned to it; Field breadth indicator QE: Analyze the number of disciplines and the degree of intersection involved in the results, use this as the basis for evaluation, and assign a score of 1-10; Then set weights for each indicator and determine their relative importance based on experience and data analysis; calculate the complexity score of scientific and technological achievement materials based on the selected indicators and weights, using the following formula: 分 =GS×ψ1+RE×ψ2+QE×ψ3, where ψ1, ψ2, and ψ3 are the weights of the technical depth index, document size index, and domain breadth index respectively; Collect historical scientific and technological achievements review data, analyze the relationship between different complexity scores and actual review time, establish a mapping function between complexity score and standard review time through data fitting method, and determine the standard review time of the scientific and technological achievements based on this function; After receiving new scientific and technological achievement materials, experts will process the tasks in order, and the efficiency of processing new tasks is the same as the efficiency of processing current tasks. The submission time of the scientific and technological achievement materials after expert review is the total time required for experts to complete all current tasks plus the standard review time of the scientific and technological achievements; According to the urgency and review cycle of the scientific and technological achievement materials, the reminder time interval is set, and reminders are issued 25%, 50% and 75% before the expected submission time respectively; according to the calculated submission time and the set reminder time interval, reminder information is automatically generated, and the reminder information includes the name of the scientific and technological achievement materials, the expert's current review progress, the estimated remaining review time and the content of the next reminder time; and it is sent to experts regularly via email, platform station message or mobile phone push to remind them to complete the review task on time.

6. The blockchain-based scientific and technological achievement authenticity verification and evaluation platform according to claim 1 is characterized in that: The process performed by the comprehensive evaluation module is as follows: Establish a comprehensive evaluation index system and assign corresponding weights to each evaluation index; these indicators include the novelty of the technology, technical difficulty, degree of improvement over existing technologies, potential market size, and environmental impact; the index weights are dynamically adjusted based on industry characteristics and market demand factors; The authenticity comprehensive value calculated by the verification module is integrated with the scores of various indicators given by the expert review; the weighted average method is used to integrate the two parts of the results according to the pre-set weights; specifically: the authenticity comprehensive value V of the verification module is integrated with the scores of various indicators given by the expert review. Assume that there are p indicators in the comprehensive evaluation indicator system, including technical novelty, expert score e1, weight a1, technical difficulty, expert score e2, weight a2, potential market size, expert score e3, weight a3; then the calculation formula of the comprehensive evaluation score S is: S = a1×e1+a2×e2+a3×e3+…+a p ×e p +b×V, where b is the weight of the authenticity comprehensive value in the comprehensive evaluation, and satisfies According to industry characteristics and market demand, we will determine a i and the value of b; At the same time, the consistency of opinions of different experts is measured by using the Kendall harmony coefficient; suppose m experts score a certain indicator of v scientific and technological achievements, and the scoring matrix is ​​R = (r ij ), where r ij represents the score of expert i on achievement j; First, calculate the score ranking of each achievement From 1 to m; then, calculate the Kendall harmony coefficient W: When W is less than the preset threshold, it is judged that the experts have large differences of opinion and the results need to be further analyzed and adjusted, including organizing expert discussions or re-evaluation; Finally, a detailed evaluation report is generated based on the comprehensive evaluation results. The report includes evaluation scores, detailed evaluation of each indicator, analysis of strengths and weaknesses, and recommendations.

Images