Published Paper Details:

The rapid spread of misinformation and fake news through online social networks has become a major societal concern, leading to political instability, social conflicts, and financial fraud. This project aims to develop a defensive machine learning model that can effectively detect and mitigate fake news propagation. The proposed system employs Natural Language Processing (NLP) techniques to analyze textual features and classify news articles as real or fake. Additionally, graph-based analysis is used to examine user behavior and detect bot accounts that contribute to the spread of misinformation. The methodology involves data collection from reliable fake news datasets (e.g., Fake Newsnet, LIAR, and Kaggle Fake News datasets), text preprocessing using TF-IDF and Word Embeddings, and classification using machine learning algorithms such as Logistic Regression, Random Forest, and LSTM (Long Short-Term Memory networks). Furthermore, network analysis using Graph Neural Networks (GNNs) and Network helps identify suspicious user activity and misinformation propagation patterns.