Hybrid Fake News Detection Model: Bagging and Logistic Regression Approach

In Today’s world or from Past Scenario’s Fake News has a Huge Impact on Everyone’s Lives including Politics, Sports, Education, External Foreign Affairs, Defense, etc. This project aims to develop a model for identifying and detecting fake news. In this model, we implement an ensemble method that leverages the TF IDF vectorization technique to extract meaningful textual features and employ a Bagging Classifier with Logistic Regression as the base estimator to enhance classification accuracy. To build this classifier, we utilized the scikit learn library. Our work was implemented using Jupyter Notebook, an online platform. For feature extraction, we employed a well-known vectorizer called Term Frequency Inverse Document Frequency (TF-IDF). We split our data into training sets (75%) and test sets (25%). The proposed model achieves an accuracy of 99.15% on the test dataset, with a precision of 99.45%, a recall of 99%, and an F1-score of 99.17%. It outperforms standalone Logistic Regression, Decision Tree, Random Forest, and Gradient Boosting classifiers.