Fuzzy Based Genetic Operators for Cyber Bullying Detection Using Social Network Data

Abstract

Social media getting more and more popular in our day today life. By the popularity of the social media affects the people who involving into it. This makes the technology to work or to feel smarter and makes us lazier. On resulting to this robust and discriminative numerical representation learning of text messages is a critical issue. Hence here we propose a learning method to tackle this issue which is named as Semantic Enhanced Marginalized Denoising Auto Encoder (smsda). Semantic extension of the popular deep learning model stacked denoising auto encoder plays a major role in this method whereas semantic extension consists of semantic dropout noise and sparsity constraints. The semantic dropout noise is designed based on domain knowledge and the word embedding technique. Our proposed method is able to exploit the hidden feature structure of bullying information and learn a robust and discriminative representation of text. Comprehensive experiments on two public cyber bullying corpora (Twitter and myspace) are conducted, and the results show that our proposed approaches outperform other baseline text representation learning methods