Deep Learning Based Fusion Approach for Hate Speech Detection

Deep Learning Based Fusion Approach for Hate Speech Detection

Pawase Shubhangi.D1, Pawar Gauri.v2, Sable Shweta.B3, Shinde Rajashree.N4 ,

Prof. Dr. Jondhale.S.R5

1,2,3,4,5Department of E&TC Engineering Amrutvahini COE, Sangamner, India

---------------------------------------------------------------------***---------------------------------------------------------------------

Abstract - Hate speech on social media has been more common in recent years, and this has drawn attention to it as a major issue across the globe. The scientific community has been interested in hate speech detection algorithms, which have received major investment from numerous governments and organisations. Although there is a wealth of literature on this topic, because each proposed solution has unique benefits and drawbacks, it is still challenging to evaluate how well it performs. A significant attempt is made to improve the classification results in general by combining the output from different classifiers. Embeddings from Language Models (ELMo), Bidirectional Encoder Representation from Transformers (BERT), and Convolutional Neural Network (CNN) are a few well-known machine learning techniques for text classification that we first examine and use.

Key Words:  machine learning, threatening speech, filtering, identification, CNN, NLP.