Volume 18, No. 6, 2021

Detect Misinformation Using Two Stage Semantic Extractor Based On Neural Network Classification

Roshan R.Karwa and Sunil R Gupta


The tremendous use of social media is the one of the important cause of generation of huge quantity of data. Analyzing this huge data is very important to get insights from the data and apply to solve real life problems. There is no accurate medium to check the semantics and authentication of data being generated. Any user of social media can post whatever they think according to their own perspection and opinion as well as user share the information without checking the authenticity, that is impacting societ in various ways. Many researchers are using Artificial Intelligence based algorithms which gives idea of detecting misinformation(commonly refereed as Fake News) potentially. Many of this techniques rely on the dataset being chosen to solve the problem. They mostly designed based on the direct feature. Understanding context with respect to its semantic is very necessary. Thus to overcome, this paper introduces Two Stage Semantic Extractor based Neural Network Method (TSENNM). According to experimental results, the proposed model obtained an good accuracy when compared with the previous model.

Pages: 3452-3464

Keywords: Artifical Intelligence, Deep Learning, Fake news, Misinformation, Neural Network. Semantic Feature extraction,

Full Text