INTERNATIONAL JOURNAL OF SCIENTIFIC DEVELOPMENT AND RESEARCH International Peer Reviewed & Refereed Journals, Open Access Journal ISSN Approved Journal No: 2455-2631 | Impact factor: 8.15 | ESTD Year: 2016
open access , Peer-reviewed, and Refereed Journals, Impact factor 8.15
HATE AND OFFENSIVE TEXT DETECTION USING DEEP LEARNING
Authors Name:
KUNDHI KIRANMAI
Unique Id:
IJSDR2309039
Published In:
Volume 8 Issue 9, September-2023
Abstract:
n recent years, people write and post abusive language on online social media platforms such as Twitter, Facebook, etc which is easily spread on internet. due to enormous volume of such posts the problem of detecting hateful and offensive text in social media is very difficult to solve manually. Hence systems that can automatically detect hate and offensive text in social media has lot of significance in modern world. In this project, Bidirectional Encoder Representation from Transformers (BERT+CNN) , Convolutional Neural Network (CNN), and Linear Short Term Memory (LSTM) are used to identify hateful text. A benchmark dataset of approximately 25 thousand annotated tweets or used to construct models based on deep learning methods. The effectiveness of BERT+CNN , CNN and LSTM models are experimentally analyzed and compared all these models classification performance. The overall aim of these project is to develop an efficient Deep Learning model for detection of hateful and offensive text automatically.
Keywords:
Enormous, Significance, CNN, LSTM, BERT, Approximately
Cite Article:
"HATE AND OFFENSIVE TEXT DETECTION USING DEEP LEARNING", International Journal of Science & Engineering Development Research (www.ijsdr.org), ISSN:2455-2631, Vol.8, Issue 9, page no.241 - 248, September-2023, Available :http://www.ijsdr.org/papers/IJSDR2309039.pdf
Downloads:
000251439
Publication Details:
Published Paper ID: IJSDR2309039
Registration ID:208496
Published In: Volume 8 Issue 9, September-2023
DOI (Digital Object Identifier):
Page No: 241 - 248
Publisher: IJSDR | www.ijsdr.org
ISSN Number: 2455-2631
Facebook Twitter Instagram LinkedIn