Design and implementation of an AI-based fake news detection system

The rapid growth of digital communication technologies and social media platforms has significantly transformed information dissemination, but has also led to the widespread proliferation of fake news, posing serious threats to public trust, social stability, and informed decision-making. Traditional methods of detecting fake news, primarily based on manual fact-checking, are increasingly inadequate due to their slow, labor-intensive, and non-scalable nature. The aim of this study is to design and implement an Artificial Intelligence (AI)-based Fake News Detection System that can automatically and efficiently classify news content as either fake or authentic. The methodology involves Natural Language Processing (NLP) techniques and Machine Learning implemented entirely in Python. Textual data is preprocessed through tokenization, stop-word removal, and lemmatization; features are then extracted using Term Frequency–Inverse Document Frequency (TF-IDF). A Logistic Regression model is trained on these features to perform classification, providing probability-based outputs that enhance interpretability. The system also incorporates a web scraping module using Requests and BeautifulSoup4 to extract content from URLs, supporting both raw text and link-based inputs. The system is deployed as a web-based application using the Flask framework. Findings reveal that the proposed system achieves reliable performance in detecting fake news with satisfactory accuracy, precision, recall, and F1-score, while maintaining computational efficiency suitable for real-time applications. It is recommended that future research extend the system to support multiple languages, incorporate deep learning models for improved accuracy, and explore multimodal approaches including images and videos.