V3I5P22

Fake News Detection System

Mrs. Subashini^1*, Needu Moijeen. M², Sameena. N³, Reshmi. S⁴     

Abstract

The Fake News Detection System is a web-based platform built using Flask that leverages machine learning to automatically classify news articles as either real or fake. The system integrates multiple functionalities: users can input text manually for prediction, access live news headlines via the NewsAPI to check their authenticity in real time, and retrieve past predictions from an SQLite database that stores the text, predicted label, confidence score, and timestamp. The machine learning model predicts the likelihood of news being fake or real, and when possible, provides a confidence score, flagging low-confidence predictions for caution. The system also exposes an API endpoint for programmatic access, allowing developers to send text in JSON format and receive predictions. By combining real-time news analysis, historical tracking, and API support, this system provides an end-to-end solution for detecting misinformation, aiding users and developers in monitoring news credibility efficiently.

Keywords:

Fake News Detection, Machine Learning, Flash, News Classification, Fake News, Real News, Web Application, NLP, News Analysis, AI Systems, SQLite Database, News Prediction, Text Analysis