What does this research mean for the field?

Transformer-based deep learning architectures provide higher prediction accuracy and better contextual understanding for fake news detection compared to traditional machine learning and other deep learning models. Novelty: ClaimNovelty.INCREMENTAL. Consensus alignment: ConsensusAlignment.NEUTRAL.

What question did this study set out to answer?

To develop an intelligent system that can automatically identify fake news using NLP and deep learning.

May 30, 2026Iconic Research and Engineering Journals0 citations

Fake News Detection using NLP and Deep Learning

PAPradip Subhash Ahire NLNiraj Mahendra Lokhande PJProf. Gadekar M. J.

Key Points

To develop an intelligent system that can automatically identify fake news using NLP and deep learning.
Applied multiple NLP preprocessing methods including tokenization and stop-word elimination.
Utilized various deep learning models such as CNN, LSTM, Bi-LSTM, and transformer-based BERT for classification.
Compared traditional machine learning algorithms with modern deep learning techniques.
Transformer-based architectures achieved higher prediction accuracy compared to traditional methods.
Deep learning models provided better contextual understanding of news articles.
The proposed system aims to improve the reliability of online content.

Abstract

The continuous growth of digital communication platforms has made information sharing easier and faster than ever before. At the same time, the spread of false and misleading news through social media, blogs, and online news portals has become a major global concern. Fake news can influence public opinion, create social panic, and negatively affect political, economic, and healthcare systems. To address this issue, researchers have focused on developing intelligent systems capable of automatically identifying fake information. This research paper presents a detailed study of fake news detection using Natural Language Processing (NLP) and Deep Learning techniques. The proposed approach applies multiple NLP preprocessing methods such as tokenization, stop-word elimination, stemming, lemmatization, and text vectorization to prepare news articles for analysis. Different deep learning models including Convolutional Neural Networks (CNN), Long Short-Term Memory (LSTM), Bidirectional LSTM (Bi-LSTM), and transformer-based BERT models are used for classification. The study also compares traditional machine learning algorithms with modern deep learning techniques. Experimental observations show that transformer-based architectures provide better contextual understanding and higher prediction accuracy. The proposed system aims to support automated misinformation detection and improve the reliability of online content.

问 AI

Bookmark

View Full Paper