Research On Trusted Thesis Review Model For Crossed Disciplines In Multiple Fields

With the continuous emergence and development of emerging disciplines and interdisciplinary disciplines,the selection of subject experts with a high degree of matching degree to review relevant scientific research papers and other discipline materials has put forward higher requirements in terms of accuracy and safety.At present,the review of papers is relatively automated,but the process is still complicated,and there are the following problems: 1.The traditional expert selection algorithm matches experts based on the subject catalog,and when facing multi-disciplinary papers,the papers cannot accurately match the review experts.2.When the education departments share papers and expert information,the privacy of the data cannot be guaranteed.3.In the process of paper review,there is a possibility of third-party tampering,which affects the credibility of the paper review data.In view of the above problems,this paper conducts a study on the trusted review process for multi-disciplinary papers,and the main work is as follows:1)A multi-disciplinary expert selection algorithm is designed,which takes the expert’s research direction keywords as the expert characteristics,and uses the Word2 vec algorithm to quantify the words of the paper and the expert’s research direction keywords.Word vector similarity between Euclidean distance calculation papers and expert research keywords based on Euclidean distance calculation papers.Based on the TF-IDF statistical value of the research keyword of the thesis,the expert matching degree is calculated in combination with the similarity of the research keywords of the paper and the expert,so as to realize the multidisciplinary expert selection.The experiment uses the dataset of 1043 experts extracted by the Aminer system,and the results show that the expert selection matching rate of the multidisciplinary expert selection algorithm reaches more than 90%.2)A multi-education center collaborative sharing structure model is constructed,and the various subjects in the network can share peer-to-peer academic data,and supervise and feedback the papers submitted by universities,expert information and the results of thesis review.The sharing model is based on the alliance chain,and the authorized nodes in the alliance chain are composed of the collection of universities of each provincial education department,and the nodes of the education department in the model realize point-to-point sharing of paper data.Experimental results show that in the multi-education center alliance chain sharing structure model,academic data can be shared between education departments safely and effectively.3)Design a trusted model for thesis review based on the consortium chain,and use the immutable nature of blockchain technology to ensure the security and credibility of thesis,expert data and thesis review process.On the basis of the collaborative sharing structure model constructed by the multi-education center,the consensus based on the multi-field and interdisciplinary expert selection algorithm is designed,and the user’s credibility measurement mechanism is set to ensure the fairness of the paper review to a certain extent.Experimental results show that the model has a good performance in data security and data transmission,which can realize the recording and storage of information in the process of paper review,and make the academic review process more secure and traceable.