Identifying Machine-Paraphrased Plagiarism. Wahle, J. P., Ruas, T., Foltýnek, T., Meuschke, N., & Gipp, B. arXiv:2103.11909 [cs], March, 2021. arXiv: 2103.11909
![Identifying Machine-Paraphrased Plagiarism [pdf]](https://bibbase.org/img/filetypes/pdf.svg "pdf")
Paper doi abstract bibtex
Paper doi abstract bibtex Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99% (F1=99.68% for SpinBot and F1=71.64% for SpinnerChief cases), while human evaluators achieved F1=78.4% for SpinBot and F1=65.6% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan. To facilitate future research, all data, code, and two web applications showcasing our contributions are openly available.
@article{WahleRFM21,
title = {Identifying {Machine}-{Paraphrased} {Plagiarism}},
url = {https://arxiv.org/pdf/2103.11909.pdf},
doi = {arXiv:2103.11909},
abstract = {Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99\% (F1=99.68\% for SpinBot and F1=71.64\% for SpinnerChief cases), while human evaluators achieved F1=78.4\% for SpinBot and F1=65.6\% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan. To facilitate future research, all data, code, and two web applications showcasing our contributions are openly available.},
urldate = {2021-04-29},
journal = {arXiv:2103.11909 [cs]},
author = {Wahle, Jan Philip and Ruas, Terry and Foltýnek, Tomáš and Meuschke, Norman and Gipp, Bela},
month = mar,
year = {2021},
note = {arXiv: 2103.11909},
keywords = {!tr\_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Digital Libraries, nlp, paraphrase, plagiarism, pre\_print, ⚠️ Invalid DOI, ⛔ No DOI found},
}Downloads: 0
{"_id":"WzyLG4KZratGco6Ce","bibbaseid":"wahle-ruas-foltnek-meuschke-gipp-identifyingmachineparaphrasedplagiarism-2021","authorIDs":["RZxtYihvstixDsXuv"],"author_short":["Wahle, J. P.","Ruas, T.","Foltýnek, T.","Meuschke, N.","Gipp, B."],"bibdata":{"bibtype":"article","type":"article","title":"Identifying Machine-Paraphrased Plagiarism","url":"https://arxiv.org/pdf/2103.11909.pdf","doi":"arXiv:2103.11909","abstract":"Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99% (F1=99.68% for SpinBot and F1=71.64% for SpinnerChief cases), while human evaluators achieved F1=78.4% for SpinBot and F1=65.6% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan. To facilitate future research, all data, code, and two web applications showcasing our contributions are openly available.","urldate":"2021-04-29","journal":"arXiv:2103.11909 [cs]","author":[{"propositions":[],"lastnames":["Wahle"],"firstnames":["Jan","Philip"],"suffixes":[]},{"propositions":[],"lastnames":["Ruas"],"firstnames":["Terry"],"suffixes":[]},{"propositions":[],"lastnames":["Foltýnek"],"firstnames":["Tomáš"],"suffixes":[]},{"propositions":[],"lastnames":["Meuschke"],"firstnames":["Norman"],"suffixes":[]},{"propositions":[],"lastnames":["Gipp"],"firstnames":["Bela"],"suffixes":[]}],"month":"March","year":"2021","note":"arXiv: 2103.11909","keywords":"!tr_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Digital Libraries, nlp, paraphrase, plagiarism, pre_print, ⚠️ Invalid DOI, ⛔ No DOI found","bibtex":"@article{WahleRFM21,\n\ttitle = {Identifying {Machine}-{Paraphrased} {Plagiarism}},\n\turl = {https://arxiv.org/pdf/2103.11909.pdf},\n\tdoi = {arXiv:2103.11909},\n\tabstract = {Employing paraphrasing tools to conceal plagiarized text is a severe threat to academic integrity. To enable the detection of machine-paraphrased text, we evaluate the effectiveness of five pre-trained word embedding models combined with machine learning classifiers and state-of-the-art neural language models. We analyze preprints of research papers, graduation theses, and Wikipedia articles, which we paraphrased using different configurations of the tools SpinBot and SpinnerChief. The best performing technique, Longformer, achieved an average F1 score of 80.99\\% (F1=99.68\\% for SpinBot and F1=71.64\\% for SpinnerChief cases), while human evaluators achieved F1=78.4\\% for SpinBot and F1=65.6\\% for SpinnerChief cases. We show that the automated classification alleviates shortcomings of widely-used text-matching systems, such as Turnitin and PlagScan. To facilitate future research, all data, code, and two web applications showcasing our contributions are openly available.},\n\turldate = {2021-04-29},\n\tjournal = {arXiv:2103.11909 [cs]},\n\tauthor = {Wahle, Jan Philip and Ruas, Terry and Foltýnek, Tomáš and Meuschke, Norman and Gipp, Bela},\n\tmonth = mar,\n\tyear = {2021},\n\tnote = {arXiv: 2103.11909},\n\tkeywords = {!tr\\_author, Computer Science - Artificial Intelligence, Computer Science - Computation and Language, Computer Science - Digital Libraries, nlp, paraphrase, plagiarism, pre\\_print, ⚠️ Invalid DOI, ⛔ No DOI found},\n}\n\n","author_short":["Wahle, J. P.","Ruas, T.","Foltýnek, T.","Meuschke, N.","Gipp, B."],"key":"WahleRFM21","id":"WahleRFM21","bibbaseid":"wahle-ruas-foltnek-meuschke-gipp-identifyingmachineparaphrasedplagiarism-2021","role":"author","urls":{"Paper":"https://arxiv.org/pdf/2103.11909.pdf"},"keyword":["!tr_author","Computer Science - Artificial Intelligence","Computer Science - Computation and Language","Computer Science - Digital Libraries","nlp","paraphrase","plagiarism","pre_print","⚠️ Invalid DOI","⛔ No DOI found"],"metadata":{"authorlinks":{"wahle, j":"https://jpelhaw.github.io/"}}},"bibtype":"article","biburl":"https://api.zotero.org/groups/2503580/items?key=t2fNBZxKJsjrlTOYPPQ4383E&format=bibtex&limit=100","creationDate":"2020-12-22T13:13:47.366Z","downloads":0,"keywords":["!tr_author","computer science - artificial intelligence","computer science - computation and language","computer science - digital libraries","nlp","paraphrase","plagiarism","pre_print","⚠️ invalid doi","⛔ no doi found"],"search_terms":["identifying","machine","paraphrased","plagiarism","wahle","ruas","foltýnek","meuschke","gipp"],"title":"Identifying Machine-Paraphrased Plagiarism","year":2021,"dataSources":["jYaxaqudzCziCrMh2"]}