Fake News Detection using LSTM Neural Network Augmented with SGD Classifier

Authors

  • Sowmya B J, Nikhil Jain C S, Seema S, K G Srinivasa

Abstract

Social media has now become the engine and outlet for untrusted news. Fake news identification is now a critical issue because it can affect the masses. A precise technique is, therefore, necessary to detect such fake news. The main aim is to detect whether a piece of information like an article is fake or not using a text-based approach Term Frequency-Inverse Document Frequency (TF-IDF) that employs deep learning techniques with the help of a Convolution Neural Network (CNN) that uses Long Short Term Memory Networks (LSTM) to learn the semantic relationships between the elements of the article and to provide category to which fake news article belongs with greater accuracy using Stochastic Gradient Descent (SGD) classifier.

Published

2020-11-01

Issue

Section

Articles