Content Based Article Authenticity Detection System
Abhishek Mitra1, Shaik Naseera2
1Abhishek Mitra, VIT University, Vellore (Tamil Nadu), India.
2Shaik Naseera, VIT University, Vellore (Tamil Nadu), India.
Manuscript received on 22 April 2019 | Revised Manuscript received on 01 May 2019 | Manuscript Published on 07 May 2019 | PP: 52-55 | Volume-7 Issue-6S3 April 2019 | Retrieval Number: F1011376S19/2019©BEIESP
Open Access | Editorial and Publishing Policies | Cite | Mendeley | Indexing and Abstracting
© The Authors. Blue Eyes Intelligence Engineering and Sciences Publication (BEIESP). This is an open access article under the CC-BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/)
Abstract: The primary objective of this project is to detect the authenticity of a news article based on the content of the same. As the classification is purely based on the article’s contents, the project uses various NLP techniques for pre-processing the text before performing classification. To make the project more user-friendly, Raspberry Pi’s Pi camera module is used to capture images of news articles which are automatically converted to text, thereby saving the hassle of typing the whole news article for the user. A comparative analysis of various machine learning and deep learning classification models is presented. This paper also presents two new approaches for article authenticity detection using deep LSTM network and deep bidirectional LSTM network. These outperform the existing approaches for detecting article authenticity and a 3.26% improvement in the F1 score from the standard existing bidirectional LSTM model is obtained.
Keywords: Artificial Neural Network; Bag of Words; Deep Learning; Fake News; Long Short Term Memory; Machine Learning; Natural Language Processing; Recurrent Neural Network; Raspberry Pi.
Scope of the Article: System Integration