Status: published
In today’s digital world, the use of online content and artificial intelligence tools has increased rapidly. Many students and content creators use AI tools to generate text or copy information from the internet, which creates challenges in maintaining originality and academic honesty. Most existing plagiarism detection systems mainly support English and are not able to effectively detect AI-generated content or work well for regional languages like Hindi and Marathi. This paper presents a review of a multilingual AI-powered system developed using Python and Flask that can detect both plagiarism and AI-generated text. The system supports English, Hindi, and Marathi languages. It uses machine learning techniques and transformer models to analyze text and determine whether it is human-written or AI-generated. For plagiarism detection, the system splits text into sentences and searches each sentence on the internet using APIs to find matching sources and calculate plagiarism percentage. The system also supports PDF file analysis by extracting text using PyMuPDF and applying both AI detection and plagiarism checking on the extracted content. Additional features such as PDF finder and text search improve the overall functionality of the system. The results are presented in a user-friendly format showing AI probability and plagiarism percentage. This system is useful for students, teachers, researchers, and content creators to ensure originality and reduce misuse of AI tools. It provides a practical and effective solution for modern challenges in content verification.
Keywords: Plagiarism Detection, AI-generated Text Detection, Natural Language Processing, Machine Learning, Multilingual System, Python, Flask, Transformer Models, PDF Processing, SerpAPI, Stylometric Analysis, Perplexity.