PasteTrace: A Single Source Plagiarism Detection Tool For Introductory Programming Courses

2025-06-20

AI Summary

Existing plagiarism detection tools (e.g., MOSS) lack fine-grained behavioral awareness and pedagogical adaptability, rendering them ineffective against covert copying—such as copy-pasting fragments from online sources—by novice programmers with weak foundational skills in introductory programming courses. Method: We propose PasteTrace, the first real-time, IDE-integrated plagiarism detector embedded in VS Code. It captures low-level coding behaviors—including copy, paste, and window-switching events—and applies temporal pattern matching alongside a lightweight heuristic rule engine to identify single-source, context-aware, fine-grained copying. Contribution/Results: Designed with pedagogical sensitivity and detection efficacy in mind, PasteTrace detects diverse copying patterns across two authentic introductory courses. It achieves higher accuracy and greater interpretability than MOSS, while enabling closed-loop instructional feedback for educators.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Humans and AI: Human-Aware Planning and Behavior Prediction
• Computer Vision: Motion & Tracking

Application Category

• Web Mining and Content Analysis: Topic discovery and tracking
• Responsible Web: Data and user privacy-enhancing technologies for the Web
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Introductory Computer Science classes are important for laying the foundation for advanced programming courses. However, students without prior programming experience may find these courses challenging, leading to difficulties in understanding concepts and engaging in academic dishonesty such as plagiarism. While there exists plagiarism detection techniques and tools, not all of them are suitable for academic settings, especially in introductory programming courses. This paper introduces PasteTrace, a novel open-source plagiarism detection tool designed specifically for introductory programming courses. Unlike traditional methods, PasteTrace operates within an Integrated Development Environment that tracks the student's coding activities in real-time for evidence of plagiarism. Our evaluation of PasteTrace in two introductory programming courses demonstrates the tool's ability to provide insights into student behavior and detect various forms of plagiarism, outperforming an existing well-established tool. A video demonstration of PasteTrace and its source code, and case study data are made available at https://doi.org/10.6084/m9.figshare.27115852

Problem

• Detects plagiarism in introductory programming courses effectively
• Tracks real-time coding activities for plagiarism evidence
• Outperforms existing tools in identifying academic dishonesty

Innovation

• Real-time coding activity tracking
• Integrated Development Environment operation
• Open-source plagiarism detection tool