A processing method and system for data element trusted circulation technology

Abstract

The present invention discloses a processing method and system for data element trusted circulation technology, which belongs to the field of data element trusted circulation technology processing. By preprocessing the circulated and trusted technology text data sets, calculating the similarity between the two and screening to obtain the trusted circulation technology text; obtaining the data transaction link and combining it with the existing key links to determine the data element circulation link; calculating the similarity between the trusted circulation technology text and the data element circulation link, classifying the trusted circulation technology text into the corresponding link, and constructing a data element trusted circulation technology system; finally, performing thematic analysis on the classified text, determining the characteristic words of the trusted circulation technology function, and then building a data element trusted circulation technology system framework; the present invention enhances the adaptability of the data element trusted circulation technology and the link, and improves the technology system framework.

Description

Technical Field

[0001] The present invention belongs to the field of data element circulation technology processing technology, and specifically relates to a data element trusted circulation technology processing method and system. Background Art

[0002] Data elements refer to data that are considered factors of production. They are data resources that can participate in social market operations and generate economic benefits for participants. Their circulation in the market can activate the value of data. During the circulation process, data elements need to be transferred between different entities. Due to ownership and security requirements, they cannot simply be transmitted directly. Trust is the cornerstone of the circulation of data elements. The "Study on Data Sharing between EU Companies" shows that approximately 73% of companies believe that technical barriers are the main obstacle to data circulation transactions. Data providers are reluctant to share data on platforms that lack technical trust, which seriously hinders the circulation of data elements in the market. Trusted circulation technology can meet the trust requirements in the circulation of data elements. A sound trusted circulation technology system can, based on the summary of current trusted circulation technologies for data elements, more comprehensively and systematically ensure the trusted circulation of data elements.

[0003] Current research on trusted data element circulation technologies primarily focuses on the application of blockchain and privacy-preserving computing technologies in various scenarios, but lacks research on technical solutions for specific aspects of the overall circulation process. Regarding data element circulation systems, research on circulation mechanisms involves relevant policies, governance systems, and business models. Research on technical implementations primarily focuses on transaction challenges, quality assessment, security oversight, and privacy-preserving computing. However, research on technical architectures lacks a comprehensive approach, with each architecture offering a single functional feature. A systematic integration guided by the circulation mechanism is needed.

[0004] The current data element trusted circulation technology system processing methods can be mainly divided into: ① Blockchain technology system: provides a distributed and transparent management, which is a trusted technology for managing digital credentials. It can realize the trusted needs in data element management, and can also solve the asset management problems faced in the circulation of data elements, and realize trusted data element asset management. Combined with blockchain, an end-to-end trust framework can be built to reduce information asymmetry in the market by bridging the trust gap; ② Privacy computing technology system: mainly represented by multi-party secure computing, federated learning and trusted execution environment. The privacy-preserving computing framework can realize the flow of data elements between multiple parties, and cross-privacy computing platform technology can realize mutual trust and intercommunication of data elements between different platforms; ③ The technical system combining blockchain and privacy technology: Blockchain and privacy technology are the main means to promote the safe and trusted circulation and transaction of data. They can help build a trusted data circulation system and a common infrastructure platform for data element circulation. The technical architecture based on blockchain and federated learning can deal with data security and privacy leakage issues in data element circulation transactions in a targeted manner.

[0005] However, the current data element trusted circulation technology has the following problems: on the one hand, it is mostly oriented towards a single circulation link and lacks technical solutions for the overall process; on the other hand, some technologies have a large research scope and do not fully consider the needs of the links in the overall process; in addition, a single technology also has defects in ensuring data credibility; in terms of technical systems, the technical architectures are diverse, each with different functional characteristics, but there is a lack of processing methods for technical systems targeting different links and specific functional requirements in the data element circulation process. Summary of the Invention

[0006] In order to solve the above problems existing in the prior art, the present invention is achieved through the following technical solutions:

[0007] In a first aspect of the present invention, a method for processing a data element trusted circulation technology is provided, comprising the following steps:

[0008] Step 1: Calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and select technical texts with a similarity greater than a set threshold as trusted circulated technical texts;

[0009] Step 2: Obtain data transaction links from the data transaction circulation links of data transaction institutions, and obtain data element circulation links based on data transaction links and existing key links of data element circulation. The data element circulation links include circulation link concepts and circulation link keywords;

[0010] Step 3: Calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system;

[0011] Step 4: Obtain the framework of the data element trusted circulation technology system based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

[0012] Step 1 includes: preprocessing the circulated technical text dataset and the trusted technical text dataset to obtain a preprocessed circulated technical text dataset and a preprocessed trusted technical text dataset, and generating a circulated technical text word vector and a trusted technical text word vector from the preprocessed circulated technical text dataset and the preprocessed trusted technical text dataset respectively;

[0013] The cosine similarity method is used to calculate the similarity between the word vectors of circulated technical texts and the word vectors of trusted technical texts, and the technical texts corresponding to the word vectors of trusted technical texts whose similarity is greater than the set threshold are selected as trusted circulated technical texts.

[0014] Screening the technical texts corresponding to the credible technical text word vectors whose similarity is greater than a set threshold as credible circulation technical texts, further comprising: traversing the initial similarity threshold, obtaining the credible circulation technical text screening accuracy based on the number of credible circulation technical texts obtained under the initial similarity threshold and the number of manually screened credible circulation technical texts, and selecting the similarity when the manually screened credible circulation technical text retention rate is 74% and the other field text elimination rate is 94% as the set threshold according to the credible circulation technical text screening accuracy.

[0015] The initial similarity threshold is greater than or equal to 0.87 and less than or equal to 0.92; the threshold is set to 0.88.

[0016] Step 3 includes: performing preliminary similarity calculation on the credible circulation technical text and the data element circulation link, and matching the credible circulation technical text and the data element circulation link based on the preliminary similarity to obtain a preliminary matching result;

[0017] The final similarity calculation is performed on the trusted circulation technical text and the data element circulation link under the preliminary matching results, and the trusted circulation technical text is classified into the data element circulation link with the highest final similarity according to the final similarity, so as to obtain the data element trusted circulation technology system.

[0018] According to the final similarity, the credible circulation technical text is classified into the circulation link of the data element with the highest final similarity, which also includes: vectorizing the credible circulation technical text, circulation link concept and circulation link keyword respectively, calculating the similarity between the vectorized credible circulation technical text and the vectorized circulation link concept and the similarity between the vectorized credible circulation technical text and the vectorized circulation link keyword respectively, and summing the similarity between the vectorized credible circulation technical text and the vectorized circulation link concept and the similarity between the vectorized credible circulation technical text and the vectorized circulation link keyword according to a certain ratio to obtain the final similarity.

[0019] Step 4 includes: performing LDA topic analysis on the trusted circulation technology text classified into the data element circulation link to obtain the topic analysis results, and obtaining the characteristic words of the trusted circulation technology function under the data element circulation link based on the topic analysis results.

[0020] The method also includes: evaluating the consistency of the topic analysis results, selecting the topic analysis results corresponding to the maximum consistency score for visualization, reducing the number of topic analysis results if there are overlapping parts in the visualization results, adjusting the adjustment parameters in the visualization, and selecting the characteristic words of the topic analysis results when the adjustment parameters are close to 0 as the characteristic words of the trustworthy circulation technical functions under the data element circulation link.

[0021] In a second aspect of the present invention, a system for processing data element trusted circulation technology is provided, which is used to execute a data element trusted circulation technology processing method provided in the first aspect of the present invention, including a text acquisition module, a data element circulation processing module, a trusted circulation system acquisition module, and a trusted circulation technology generation module;

[0022] The text acquisition module is used to calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and select technical texts with similarity greater than a set threshold as trusted circulated technical texts;

[0023] The data element circulation processing module is used to obtain data transaction links from the data transaction circulation links of the data transaction institutions, and obtain the data element circulation links according to the data transaction links and the existing key links of data element circulation. The data element circulation links include the concept of circulation links and circulation link keywords;

[0024] The trusted circulation system acquisition module is used to calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system;

[0025] The trusted circulation technology generation module is used to obtain the data element trusted circulation technology system framework based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

[0026] Compared with the prior art, the present invention has the following beneficial effects:

[0027] 1. The present invention calculates the similarity between the circulated technical text dataset and the trusted technical text dataset, screens out the trusted circulated technical text and matches it with the data element circulation link, constructs a trusted circulation technology system that runs through the entire process of data element circulation, integrates the technologies of each link, and forms a complete technical system framework, overcoming the problem that the existing technology only focuses on the local and lacks overall planning, so that data elements can get effective technical support throughout the entire circulation process.

[0028] 2. When matching technical texts with links, the present invention vectorizes the concepts and keywords of the circulation links respectively, and comprehensively calculates the similarity with the credible circulation technical texts, fully considering the specific characteristics and requirements of each link, improving the adaptability of technology and links, and ensuring that the solutions in the technical texts can better serve all links of data element circulation.

[0029] 3. The present invention classifies the trusted circulation technology texts into the corresponding data element circulation links through similarity calculation, constructs a data element trusted circulation technology system, and performs LDA topic analysis on the trusted circulation technology texts classified into the data element circulation links. According to the topic analysis results, the characteristic words of the trusted circulation technology functions are determined, and the technical system framework is further improved, so that the technical system can be constructed according to the functional requirements of different links, and the various architectures cooperate with each other, and the functions are more comprehensive and rich, which makes up for the shortcomings of the existing technical system architecture.

[0030] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0031] Figure 1 This is a flow chart of a processing method for a data element trusted circulation technology provided by an embodiment of the present invention;

[0032] Figure 2 This is a data element circulation link diagram provided by an embodiment of the present invention;

[0033] Figure 3 This is a framework diagram of a data element trusted circulation technology system provided by an embodiment of the present invention;

[0034] Figure 4 It is a structural diagram of a processing system for a data element trusted circulation technology provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0035] In order to further illustrate the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, the scheme according to the present invention is described in detail below with reference to the accompanying drawings and specific implementation methods.

[0036] The aforementioned and other technical contents, features, and effects of the present invention are clearly presented in the following detailed description of the specific embodiments in conjunction with the accompanying drawings. Through the description of the specific embodiments, a deeper and more specific understanding of the technical means and effects adopted by the present invention to achieve the intended purpose can be obtained. However, the accompanying drawings are provided for reference and illustration purposes only and are not intended to limit the technical solutions of the present invention.

[0037] It should be noted that, in this document, relational terms such as first and second, etc., are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprises," "comprising," or any other variations are intended to cover non-exclusive inclusion, such that an article or device comprising a series of elements includes not only those elements, but also other elements not explicitly listed. In the absence of further limitations, an element defined by the phrase "comprising a ..." does not exclude the presence of additional identical elements in the article or device comprising the element.

[0038] like Figure 1 As shown, a method for processing data element trusted circulation technology provided by an embodiment of the present invention includes the following steps:

[0039] Step 1: Calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and select technical texts with similarity greater than the set threshold as trusted circulated technical texts.

[0040] Texts containing relevant technical requirements in the circulation process of data elements were screened out from existing policy texts as the circulated technology text dataset; trusted technology-related literature with the theme of "trusted technology" was searched on a core journal website, and trusted technology-related patents with the abstract of "trusted technology" were searched on a patent website. The trusted technology text dataset is composed of trusted technology-related literature and trusted technology-related patents.

[0041] The circulated technical text dataset and the trusted technical text dataset were preprocessed using the zh_core_web_md Chinese model in the open source NLP natural language processing library spaCy to obtain the preprocessed circulated technical text dataset and the preprocessed trusted technical text dataset. Based on the Word2Vec algorithm, the preprocessed circulated technical text dataset and the preprocessed trusted technical text dataset were used to generate circulated technical text word vectors and trusted technology text word vectors respectively.

[0042] Preprocessing the dataset using spaCy's zh_core_web_md Chinese model can save time and computing resources, and can accurately identify various linguistic phenomena, providing high-quality data for subsequent tasks; generating tradable technical text word vectors and trusted technical text word vectors through the Word2Vec algorithm can capture the semantic relationship between words and perceive the context, accurately understand the meaning of terms in technical texts in different contexts, improve the accuracy of semantic understanding, and convert technical texts into numerical and low-dimensional word vectors, significantly improving the processing efficiency of large-scale technical text datasets.

[0043] The cosine similarity method is used to calculate the similarity between the word vectors of circulated technical texts and the word vectors of trusted technical texts, and the technical texts corresponding to the word vectors of trusted technical texts whose similarity is greater than the set threshold are selected as trusted circulated technical texts.

[0044] The screening accuracy of credible circulating technical texts is obtained based on the number of credible circulating technical texts obtained under the initial similarity threshold and the number of credible circulating technical texts screened manually. The calculation method is:

[0045] .

[0046] in, Precision Screening accuracy for trusted circulation technical texts, TP is the number of credible circulating technical texts obtained under the initial similarity threshold and the number of duplicate credible circulating technical texts screened manually, FP It is the number of credible circulating technical texts in other fields obtained under the initial similarity threshold.

[0047] When the similarity threshold is less than 0.87, the number of credible circulation technical texts obtained exceeds the number of manually screened ones; when the similarity threshold is greater than 0.92, the number of credible circulation technical texts obtained is 0, therefore, the initial similarity threshold is greater than or equal to 0.87 and less than or equal to 0.92; the initial similarity threshold is traversed, and the credible circulation technical text screening accuracy is obtained according to the number of credible circulation technical texts obtained under the initial similarity threshold and the number of manually screened credible circulation technical texts. According to the credible circulation technical text screening accuracy, the similarity when the manually screened credible circulation technical text retention rate is high and the rejection rate of texts in other fields is high is selected as the set threshold. In this embodiment, the manually screened credible circulation technical text retention rate is as high as 74%, and the rejection rate of texts in other fields is as high as 94%. At this time, the threshold is set to 0.88.

[0048] Step 2: Obtain the data transaction links from the data transaction circulation links of the data transaction institutions, and obtain the data element circulation links based on the data transaction links and the existing key links of data element circulation. The data element circulation links include the circulation link concept and circulation link keywords.

[0049] A data trading institution with a data trading circulation link is selected from existing data trading institutions. Since the data trading link is an important link in the data trading circulation link, the data trading link can be obtained based on the data trading circulation link.

[0050] like Figure 2 As shown in FIG. 1 , a data element circulation link provided by an embodiment of the present invention is provided, wherein the concept of the circulation link includes:

[0051] The key words for the circulation link are data collection and data aggregation of multi-source heterogeneous data access.

[0052] The key words for the circulation link are data cleaning, data transformation, data integration, data desensitization, data reduction, data labeling, data encryption, improving data quality, automatic quality verification, data processing and data conversion.

[0053] The key words for the circulation link are data storage, data backup, database management, data management, data recovery, encrypted storage and data hosting with storage isolation.

[0054] The keywords for the circulation link are data product development, including data modeling, data mining, data computing, data products and data analysis.

[0055] The key words for the circulation link are data product registration and evaluation of content registration, ownership registration, ownership declaration, registration certification, data notarization, product disclosure, asset evaluation, quality evaluation, compliance evaluation, title confirmation evaluation, information storage, product traceability and product compliance review.

[0056] The key words for the circulation link are data product listing, product pre-release, data product query, demand release, transaction matching, transaction signing, transaction evaluation, transaction verification and data product transaction and use of agreement delivery.

[0057] Data governance includes the strategies, standards and processes for the circulation of data elements to ensure the quality, security, compliance and effectiveness of the circulation of data elements; transaction compliance supervision includes the systems and norms for the circulation of data elements to ensure the circulation of data elements from the aspects of management systems and norms.

[0058] Step 3: Calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system.

[0059] The Sentence-Bert algorithm is used to calculate the preliminary similarity between the trusted circulation technical text and the data element circulation link, and the trusted circulation technical text and the data element circulation link are matched according to the preliminary similarity to obtain the preliminary matching results.

[0060] The terms in the trusted circulation technical text and the data element circulation link will have different meanings depending on the context. The Sentence-Bert algorithm can effectively identify these differences and improve the accuracy of similarity calculation. At the same time, it can ignore the interference factors caused by non-standard expressions in the descriptions of the trusted circulation technical text and the data element circulation link, thereby ensuring the stability of the similarity calculation.

[0061] The trusted circulation technology text and data element circulation link based on the preliminary matching results are vectorized respectively. The vectorization of the data element circulation link includes the vectorization of the circulation link concept and the vectorization of the circulation link keywords. The paraphrase-multilingual-MiniLM-L12-v2 model is used to calculate the similarity between the vectorized trusted circulation technology text and the vectorized circulation link concept (concept_similarities) and the similarity between the vectorized trusted circulation technology text and the vectorized circulation link keywords (keywords_similarities). The similarity between the vectorized trusted circulation technology text and the vectorized circulation link concept and the similarity between the vectorized trusted circulation technology text and the circulation link keywords are summed according to a certain ratio to obtain the final similarity. Among them, the best final similarity is obtained when the concept_similarities account for 30% and the keywords_similarities account for 70%. According to the final similarity, the trusted circulation technology text is classified into the data element circulation link with the highest final similarity to obtain the data element trusted circulation technology system.

[0062] paraphrase-multilingual-MiniLM-L12-v2 is a model that can map sentences and paragraphs into a 384-dimensional dense vector space. It supports multiple languages, is lightweight and efficient, has strong semantic representation capabilities, and is accurate in context awareness. It also has good generalization and robustness. It can reliably process vectorized trusted circulation technical text and vectorized data element circulation links, and derive the similarity between them.

[0063] Step 4: Obtain the framework of the data element trusted circulation technology system based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

[0064] Perform LDA topic analysis on the trusted circulation technology texts classified into the data element circulation link according to the final similarity in step 3 to obtain the topic analysis results, and use the cv consistency score to evaluate the consistency of the topic analysis results. Select the topic analysis results corresponding to the maximum cv consistency score for pyLDAvis visualization. If there are overlapping parts in the pyLDAvis visualization results, reduce the number of topic analysis results, and adjust the adjustment parameter λ in the pyLDAvis visualization so that the adjustment parameter λ approaches 0. Select the topic analysis result feature words when the adjustment parameter λ approaches 0 as the feature words of the trusted circulation technology function under the data element circulation link. The topic analysis result feature words are preferably the top 10 feature words.

[0065] The closer the adjustment parameter λ is to 1, the more relevant the frequently appearing words in the topic analysis results are to the topic; the closer the adjustment parameter λ is to 0, the more relevant the more special and unique words in the topic analysis results are to the topic. By adjusting the size of λ, the occurrence results of feature words can be changed. In order to better express the uniqueness of different topic analysis results in the data element circulation link, the λ value is as close to 0 as possible. Preferably, the adjustment parameter λ is 0.6.

[0066] The use of LDA topic analysis can automatically mine the potential topic structure of credible circulation technology texts, achieve dimensionality reduction and information compression, and reveal topic associations. At the same time, LDA topic analysis can adapt to the diversity of credible circulation technology texts; and topic consistency is an important indicator for evaluating the coherence and interpretability of topic analysis results. It can measure the similarity between the top N words with the highest probability in each topic analysis result. The coherence of the topic analysis results can be quantified through the cv consistency score, providing an intuitive numerical measurement of the similarity between the top N words with the highest probability in each topic analysis result.

[0067] According to the characteristic words of the trusted circulation technology function under the data element circulation link and the trusted circulation technology system obtained in step 3, the data element trusted circulation technology system framework is obtained, such as Figure 3 As shown, a data element trusted circulation technology system framework provided by an embodiment of the present invention is as follows:

[0068] The data governance and transaction compliance supervision on the left are the systems and norms for the corresponding links in the circulation of intermediate data elements; the intermediate data element circulation links are organized from bottom to top according to the data element circulation process, including the data aggregation link composed of data security and traceability system and data intelligent integration and detection; the data processing link composed of data processing and data conversion and traceability under quality monitoring; the data hosting link composed of user services and device management, privacy protection and encrypted storage, blockchain-based traceability and auditing, and access control and identity authentication; the data product development link composed of data model construction, intelligent representation and analysis, product development and data services; the data product registration and evaluation link composed of data asset evaluation, data analysis and authentication, and data rating and market access authorization; the product data transaction and use link composed of blockchain transaction system, transaction platform service management, transaction credit points model, product information management and transaction authentication; the data transmission on the right includes transmission security and privacy protection and distributed information storage.

[0069] like Figure 4 As shown, a processing system for data element trusted circulation technology provided by an embodiment of the present invention is used to execute the processing method for data element trusted circulation technology of the present invention, including a text acquisition module, a data element circulation processing module, a trusted circulation system acquisition module and a trusted circulation technology generation module.

[0070] The text acquisition module is used to calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and screen the technical texts with similarity greater than the set threshold as trusted circulated technical texts.

[0071] The data element circulation processing module is used to obtain data transaction links from the data transaction circulation links of data transaction institutions, and obtain data element circulation links based on data transaction links and existing data element circulation key links. The data element circulation links include circulation link concepts and circulation link keywords.

[0072] The trusted circulation system acquisition module is used to calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system.

[0073] The trusted circulation technology generation module is used to obtain the data element trusted circulation technology system framework based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

[0074] The present invention preprocesses a data set of circulated and trusted technical texts, generates word vectors using the Word2Vec algorithm, and selects trusted circulation technical texts through cosine similarity calculation and threshold adjustment; obtains data transaction links from data transaction institutions, and derives data element circulation links in combination with existing key links; then uses the Sentence-Bert algorithm and the paraphrase-multilingual-MiniLM-L12-v2 model to calculate the similarity between the two, classifies the trusted circulation technical texts, and forms a trusted circulation technology system; performs LDA topic analysis on the classified texts, optimizes the topic analysis results with the help of CV consistency scores to determine feature words, and obtains a trusted circulation technology system framework, which overcomes the problem of the existing technology that only focuses on local areas and lacks overall planning, improves the adaptability of data element trusted circulation technologies and links, improves the technical system framework, makes the data element trusted circulation technology system more comprehensive and rich in functions, and enhances the trust level of data circulation transactions.

[0075] The above is a further detailed description of the present invention in conjunction with specific preferred embodiments, and the specific implementation of the present invention should not be considered to be limited to these descriptions. For those skilled in the art of the present invention, without departing from the concept of the present invention, several simple deductions or substitutions can be made, which should be considered to fall within the scope of protection of the present invention.

Claims

1. A method for processing data element trusted circulation technology, characterized in that: The following steps are involved: Step 1: Calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and select technical texts with a similarity greater than a set threshold as trusted circulated technical texts; Step 2: Obtain data transaction links from the data transaction circulation links of data transaction institutions, and obtain data element circulation links based on data transaction links and existing key links of data element circulation. The data element circulation links include circulation link concepts and circulation link keywords; Step 3: Calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system; Step 4: Obtain the framework of the data element trusted circulation technology system based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

2. A method for processing data element trusted circulation technology according to claim 1, characterized in that: The step 1 comprises: Preprocessing the circulated technical text dataset and the trusted technical text dataset respectively to obtain a preprocessed circulated technical text dataset and a preprocessed trusted technical text dataset, and generating circulated technical text word vectors and trusted technical text word vectors from the preprocessed circulated technical text dataset and the preprocessed trusted technical text dataset respectively; The cosine similarity method is used to calculate the similarity between the word vectors of circulated technical texts and the word vectors of trusted technical texts, and the technical texts corresponding to the word vectors of trusted technical texts whose similarity is greater than the set threshold are selected as trusted circulated technical texts.

3. A method for processing data element trusted circulation technology according to claim 2, characterized in that: The step of screening the technical texts corresponding to the credible technical text word vectors whose similarity is greater than a set threshold as credible circulating technical texts further includes: Traverse the initial similarity threshold, and obtain the screening accuracy of credible circulation technical texts based on the number of credible circulation technical texts obtained under the initial similarity threshold and the number of manually screened credible circulation technical texts. According to the screening accuracy of credible circulation technical texts, the similarity when the manually screened credible circulation technical text retention rate is 74% and the elimination rate of texts in other fields is 94% is selected as the set threshold.

4. A method for processing data element trusted circulation technology according to claim 3, characterized in that: The initial similarity threshold is greater than or equal to 0.87 and less than or equal to 0.92; the set threshold is 0.

88.

5. The method for processing data element trusted circulation technology according to claim 1, characterized in that: The step 3 comprises: Performing preliminary similarity calculation on the credible circulation technical text and the data element circulation link, and matching the credible circulation technical text and the data element circulation link based on the preliminary similarity to obtain a preliminary matching result; The final similarity calculation is performed on the trusted circulation technical text and the data element circulation link under the preliminary matching results, and the trusted circulation technical text is classified into the data element circulation link with the highest final similarity according to the final similarity, so as to obtain the data element trusted circulation technology system.

6. A method for processing data element trusted circulation technology according to claim 5, characterized in that: The categorization of the credible circulation technical text into the data element circulation link with the highest final similarity according to the final similarity also includes: The trusted circulation technology text, circulation link concepts and circulation link keywords are vectorized respectively, and the similarity between the vectorized trusted circulation technology text and the vectorized circulation link concepts and the similarity between the vectorized trusted circulation technology text and the vectorized circulation link keywords are calculated respectively. The similarity between the vectorized trusted circulation technology text and the vectorized circulation link concepts and the similarity between the vectorized trusted circulation technology text and the vectorized circulation link keywords are summed up according to a certain ratio to obtain the final similarity.

7. A method for processing data element trusted circulation technology according to claim 1, characterized in that: The step 4 comprises: LDA topic analysis is performed on the trusted circulation technology texts classified into the data element circulation link to obtain the topic analysis results, and the characteristic words of the trusted circulation technology function under the data element circulation link are obtained based on the topic analysis results.

8. A method for processing data element trusted circulation technology according to claim 7, characterized in that: The characteristic words of the trusted circulation technology function in the data element circulation link obtained according to the subject analysis results also include: Evaluate the consistency of the topic analysis results, select the topic analysis results corresponding to the maximum consistency score for visualization, reduce the number of topic analysis results if there are overlapping parts in the visualization results, and adjust the adjustment parameters in the visualization, and select the characteristic words of the topic analysis results when the adjustment parameters are close to 0 as the characteristic words of the trusted circulation technology function under the data element circulation link.

9. A data element trusted circulation technology processing system, characterized in that: A method for processing a data element trusted circulation technology according to any one of claims 1 to 8, comprising a text acquisition module, a data element circulation processing module, a trusted circulation system acquisition module, and a trusted circulation technology generation module; The text acquisition module is used to calculate the similarity between the circulated technical text dataset and the trusted technical text dataset, and select technical texts with similarity greater than a set threshold as trusted circulated technical texts; The data element circulation processing module is used to obtain data transaction links from the data transaction circulation links of the data transaction institution, and obtain the data element circulation links according to the data transaction links and the existing data element circulation key links. The data element circulation links include circulation link concepts and circulation link keywords; The trusted circulation system acquisition module is used to calculate the similarity between the trusted circulation technical text and the data element circulation link, and classify the trusted circulation technical text into the data element circulation link with the highest similarity to obtain the data element trusted circulation technical system; The trusted circulation technology generation module is used to obtain the data element trusted circulation technology system framework based on the characteristic words of the trusted circulation technology function under the data element circulation link and the data element trusted circulation technology system.

Images