Comparative Analysis of Machine Learning Algorithms for Detection of Online Hate Speech
Volume 18, Issue 2, 2024
Download| Author(s): |
Atiya Masood Iqra University, Karachi, Pakistan, atiya.masood@iqra.edu.pk Syed Muhammad Daniyal* Iqra University, Karachi, Pakistan, syed.daniyal@iqra.edu.pk Burhan Ahmed Khanzada Iqra University, Karachi, Pakistan, burhan.khanzada@outlook.com Uzair Aftab Iqra University, Karachi, Pakistan, uzair.aftab@outlook.com |
|---|---|
| Abstract | Hate speech is a severe problem affecting online social communities' functionality and dynamics. The importance of developing machine-learning techniques for hate speech identification has risen. Moreover, the literature reported that supervised learning performs better in detecting and classifying hateful content. This study compares the performance of several supervised approaches for detecting online hate speech. This study looks closely at the methodologies and their generalizability to existing hate speech datasets. For this purpose, we used several datasets for our experiments and performed text processing, feature engineering, data splitting, and classifier evaluation. We assessed the performance of logistic regression, decision tree, random forest, naive Bayes, and light gradient boosting machine (LGBM) algorithms using the Bag of Words (BoW) and Term Frequency-Inverse Document Frequency (TF-IDF) feature extraction techniques, and our findings confirm known bias in the dataset and methodological issues. In-depth, we have mainly found issues with sampling and data over fitting. |
| Keywords | Hate Speech Detection, Machine Learning, Supervised Learning, Generalizability, Text Processing |
| Year | 2024 |
| Volume | 18 |
| Issue | 2 |
| Type | Research paper, manuscript, article |
| Journal Name | Journal of Information & Communication Technology | Publisher Name | ILMA University | Jel Classification | - | DOI | - | ISSN no (E, Electronic) | 2075-7239 | ISSN no (P, Print) | 2415-0169 | Country | Pakistan | City | Karachi | Institution Type | University | Journal Type | Open Access | Manuscript Processing | Blind Peer Reviewed | Format | Paper Link | https://jict.ilmauniversity.edu.pk/journal/jict/18.2/1.pdf | Page |