Plagiarism is a concern in educational institutions because students may use existing work without giving proper credit. Detecting rewritten text and processing scanned or handwritten documents can be challenging for plagiarism detection methods. This paper proposes an AI-based plagiarism detection system that combines Natural Language Processing (NLP) deep learning and document text extraction. The system provides portals for students and teachers. Students can submit documents scanned files and handwritten assignments in supported formats. Text extraction and Optical Character Recognition (OCR) are used to convert document content into text for analysis. The extracted text is examined using matching and semantic similarity techniques to identify possible similarities with reference documents and previous student submissions. The proposed system also provides teacher dashboards for reviewing student records submitted assignments and plagiarism reports. These reports are intended to help teachers examine matching passages and assess the submitted content. The systems performance will be evaluated using accuracy, precision, recall and F1-score. The main objective is to support integrity by improving the analysis of textual similarities including cases where the wording differs but the meaning remains similar.

Prof. Shital Patil, Komal Sonawane, Sakshi Jaiswal, Sanskar Sonawane, Karishma Verma, "AI-Based Plagiarism Detection System Using Deep Learning" International Journal of Computer Science and Engineering Development, V1(4): Page(243-245) September - October 2026. ISSN: 3139-0862. www.ijcsed.com. Published by Scientific and Academic Research Publishing.