Design And Implementation Of A Cloud Platform For Plagiarism Check In English Writing

English composition is one of the important indicators to measure the comprehensive English level of Chinese students.It is not only a comprehensive evaluation of students’ English vocabulary and grammar mastery,but also a specific reflection of students’ comprehensive ability in choosing words,making sentences,and embellishing rhetoric.However,in the actual teaching and examination of English writing,plagiarism emerges endlessly due to factors such as some students’ misbehavior and teachers’ poor supervision.In response to the above situation,researchers at home and abroad have carried out a lot of research,divided students’ plagiarism into copy-paste plagiarism and text rewriting plagiarism,and tried both kinds of plagiarism.However,most of the existing models still have problems such as single inspection,slow inspection speed,and unsatisfactory plagiarism inspection results,and cannot cope well with writing plagiarism inspection.Based on the above problems,the paper designs a copy-and-paste plagiarism check model based on digital fingerprints and a text rewriting plagiarism check model based on deep learning.The models have improved in terms of checking speed and checking accuracy.In addition,in order to meet the actual needs of plagiarism checks in English writing,a cloud platform for plagiarism checks in English writing that serves as a unit has been implemented.The main research content of the paper is as follows:(1)Aiming at the copy-and-paste situation in English writing plagiarism,a digital fingerprint model based on N-Gram window jumping mechanism is designed.Based on digital fingerprint technology,this model adds sliding window and improved matching mechanism,which solves the problems of excessive fingerprint density and low checking efficiency in text feature extraction.At the same time,the model adds the Fisher–Yates shuffle algorithm with a salt parameter,which solves the problems caused by hash collisions in the process of text comparison.Experiments show that the model can effectively detect copy-paste plagiarism in English compositions in a short time.(2)Aiming at the text rewriting in English writing plagiarism,a fusion model based on Text CNN-Bi GRU is designed,which combines the convolutional neural network-based text classification model(Text CNN)and the bidirectional gated recurrent unit(Bi GRU),so that the extracted text semantic information can take into account both text local features and text context features,so as to detect text rewriting more accurately.In the experiment,the model uses the MRPC data set as the experimental data,explores the influence of different word vector dimensions,iteration times and other parameter variables on the inspection effect,and compares it with other models in related fields.The experimental results show that the model is better than WMD,DSSM,CDSSM and other models improve the accuracy by 1.9 to 11.2percentage points,and the F1 value by 2.0 to 15.3 percentage points.(3)On the basis of the proposed English writing plagiarism checking model,a cloud platform architecture with high concurrency,non-blocking,and safe multi-file upload/download between the client and the server is designed.This architecture uses the C/S mode,based on the distributed server architecture,and through the custom network protocol,it realizes the fast and stable plagiarism check for the English writing of middle school students in the class.Experimental results from protocol testing and stress testing show that this architecture has improved request response speed and file transfer performance compared to traditional server architectures.