Summary of Bert-enhanced Retrieval Tool For Homework Plagiarism Detection System, by Jiarong Xian et al.
BERT-Enhanced Retrieval Tool for Homework Plagiarism Detection System
by Jiarong Xian, Jibao Yuan, Peiwei Zheng, Dexian Chen, Nie yuntao
First submitted to arxiv on: 1 Apr 2024
Categories
- Main: Computation and Language (cs.CL)
- Secondary: Artificial Intelligence (cs.AI); Information Retrieval (cs.IR)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary This paper proposes a method for generating plagiarized text datasets based on GPT-3.5, addressing the lack of high-quality datasets for detecting high-level plagiarism in natural language processing (NLP). The generated dataset consists of 32,927 pairs of texts covering various plagiarism methods, making it a valuable resource for researchers. Additionally, the paper presents a plagiarism identification method using Faiss with BERT, achieving high efficiency and accuracy. Experimental results show that this model outperforms others in metrics such as accuracy (98.86%), precision (98.90%), recall (98.86%), and F1 score (0.9888). The proposed demo platform allows users to upload a text library and participate in plagiarism analysis. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary This paper helps solve a big problem in computer science: detecting when someone copies and pastes from another source without giving credit. Right now, we don’t have many good examples of texts that show this kind of copying, making it hard to develop better ways to detect plagiarism. This research creates a large collection of fake text pairs that demonstrate different types of copying, which will help researchers make progress on this important task. The team also developed a way to quickly and accurately identify plagiarized texts using a special combination of computer models. They tested their method and showed it works well, even better than some other methods. Finally, they created a user-friendly platform that lets anyone test their own text library for plagiarism. |
Keywords
» Artificial intelligence » Bert » F1 score » Gpt » Natural language processing » Nlp » Precision » Recall
