Plagiarism-Detection

Using the BERT (Bidirectional Encoder Representations from Transformers) model, users can compare two documents in this Streamlit-based web tool to identify possible instances of plagiarism. For document comparison, the program allows for both file uploads and text input.

Features

• Text Input: Manually input text into the app for plagiarism detection.
• File Upload: Upload .txt, .pdf, or .docx files for plagiarism detection.
• Cosine Similarity: Calculates similarity between documents using cosine similarity.
• Threshold Adjustment: Allows users to set a similarity threshold for detecting plagiarism.

Install Dependencies

pip install -r requirements.txt

Running the App Locally

streamlit run Plagiarism_Detector.py