Please login first
Detection of misogyny from Arabic Levantine Twitter tweets using machine learning techniques
1, 2 , 3, 4 , 5 , 5 , * 6
1  Department of Studies in Computer Science, University of Mysore , Manasagangothri, Mysore 570006, India
2  Sana’a Community College, Sana’a 5695, Yemen
3  Department of Computer Science and Engineering, College of Software, Kyung Hee University, 17104 Gyeonggi-do, Republic of Korea;
4  Sana’a Community College, 5695 Sana'a, Republic of Yemen
5  Dept.of Computer science and , Indian Institute of InformationTechnology, Kottayam, India
6  Dept.of Studies in Computer Science, University of Mysore, Mysuru, India
Academic Editor: Frank Werner

Published: 19 September 2021 by MDPI in The 1st Online Conference on Algorithms session Artificial Intelligence Algorithms
Abstract:

Due to the increased abundance of user comments on the internet, such as Twitter, Arabic Text Detection (ATD) is one of the most difficult computational tasks for the machine learning field. Misogyny in Arabic text detection has become a touchy issue, especially among Arab women. Online misogyny has become a major threat to women in many countries, and online misogynistic harassment has grown in recent years In this article, we use misogynistic women in Levantine as a case study to build a new approach for detecting Arabic text. The suggested study's goal is to discover a novel Arabic text recognition algorithm for misogyny of women in Arabic countries. Our approach has been evaluated on the Arabic Levantine Twitter Dataset for Misogynistic, and we achieved an excellent accuracy of 90% using the BERTv2n in binary classification and 89 in multi classification .

Keywords: Arabic language; Text pre-processing; Representation; Text Detection Technique, Misogyny of women.
Top