Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms

Currently social media and online platforms have become a major source of cyberbullying and hate speech. It is currently affecting people and communities in harmful ways. Hate speech on social media is rising in Bangladesh and it is creating a need for effective tools t...

Deskribapen osoa

Gorde:
Xehetasun bibliografikoak
Egile Nagusiak: Islam Arif, Md Ariful, Rahman, Md. Mahbubur, Rabiul Alam, Md. Golam, Akhtaruzzaman, M.
Formatua: Artikulua
Hizkuntza:ingelesa
Argitaratua: Research and Development Wing, MIST 2025
Gaiak:
Sarrera elektronikoa:http://dspace.mist.ac.bd:8080/xmlui/handle/123456789/865
Etiketak: Etiketa erantsi
Etiketarik gabe, Izan zaitez lehena erregistro honi etiketa jartzen!
_version_ 1868226992644554752
author Islam Arif, Md Ariful
Rahman, Md. Mahbubur
Rabiul Alam, Md. Golam
Akhtaruzzaman, M.
author_browse Akhtaruzzaman, M.
Islam Arif, Md Ariful
Rabiul Alam, Md. Golam
Rahman, Md. Mahbubur
author_facet Islam Arif, Md Ariful
Rahman, Md. Mahbubur
Rabiul Alam, Md. Golam
Akhtaruzzaman, M.
author_sort Islam Arif, Md Ariful
collection DSpace
description Currently social media and online platforms have become a major source of cyberbullying and hate speech. It is currently affecting people and communities in harmful ways. Hate speech on social media is rising in Bangladesh and it is creating a need for effective tools to prevent and detect these incidents. This study introduces a deep learning model to mitigate this issue of identifying hate speech in text using three types of word embedding methods: Word2Vec, FastText, and BERT. The text data was labeled to mark hate speech and non-hate speech content. After that, these texts are preprocessed by removing punctuation and symbols to help improve model accuracy. Five deep learning models Bi-GRU-LSTM-CNN, Bi-LSTM, CNN, LSTM, and XGBoost were trained to classify the text as hate speech or non-hate speech. The study found that the LSTM model accomplished the highest accuracy at 95.66% with the Word2Vec embedding method, while CNN reached 87.70% with FastText embeddings. Word2Vec is effective for capturing word meanings in general text classification. FastText works well with rare words and languages that have complex word forms. These findings help advance effective hate speech detection techniques. It could promote more respectful and inclusive interactions on social media. This proposed deep-learning model can help stop cyberbullying and hate speech on social media.
format Article
id oai:localhost:123456789-865
institution My University
language English
publishDate 2025
publishDateRange 2025
publishDateSort 2025
publisher Research and Development Wing, MIST
publisherStr Research and Development Wing, MIST
record_format dspace
spelling oai:localhost:123456789-8652025-05-07T03:23:48Z Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms Islam Arif, Md Ariful Rahman, Md. Mahbubur Rabiul Alam, Md. Golam Akhtaruzzaman, M. Social media platform, Hate speech detection, Deep learning models, Word embedding, LSTM Currently social media and online platforms have become a major source of cyberbullying and hate speech. It is currently affecting people and communities in harmful ways. Hate speech on social media is rising in Bangladesh and it is creating a need for effective tools to prevent and detect these incidents. This study introduces a deep learning model to mitigate this issue of identifying hate speech in text using three types of word embedding methods: Word2Vec, FastText, and BERT. The text data was labeled to mark hate speech and non-hate speech content. After that, these texts are preprocessed by removing punctuation and symbols to help improve model accuracy. Five deep learning models Bi-GRU-LSTM-CNN, Bi-LSTM, CNN, LSTM, and XGBoost were trained to classify the text as hate speech or non-hate speech. The study found that the LSTM model accomplished the highest accuracy at 95.66% with the Word2Vec embedding method, while CNN reached 87.70% with FastText embeddings. Word2Vec is effective for capturing word meanings in general text classification. FastText works well with rare words and languages that have complex word forms. These findings help advance effective hate speech detection techniques. It could promote more respectful and inclusive interactions on social media. This proposed deep-learning model can help stop cyberbullying and hate speech on social media. 2025-05-07T03:23:47Z 2025-05-07T03:23:47Z 2024-12 Article 2224-2007 http://dspace.mist.ac.bd:8080/xmlui/handle/123456789/865 en application/pdf Research and Development Wing, MIST
spellingShingle Social media platform, Hate speech detection, Deep learning models, Word embedding, LSTM
Islam Arif, Md Ariful
Rahman, Md. Mahbubur
Rabiul Alam, Md. Golam
Akhtaruzzaman, M.
Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title_full Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title_fullStr Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title_full_unstemmed Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title_short Analyzing the Performance of Deep Learning Models for Detecting Hate Speech on Social Media Platforms
title_sort analyzing the performance of deep learning models for detecting hate speech on social media platforms
topic Social media platform, Hate speech detection, Deep learning models, Word embedding, LSTM
url http://dspace.mist.ac.bd:8080/xmlui/handle/123456789/865
work_keys_str_mv AT islamarifmdariful analyzingtheperformanceofdeeplearningmodelsfordetectinghatespeechonsocialmediaplatforms
AT rahmanmdmahbubur analyzingtheperformanceofdeeplearningmodelsfordetectinghatespeechonsocialmediaplatforms
AT rabiulalammdgolam analyzingtheperformanceofdeeplearningmodelsfordetectinghatespeechonsocialmediaplatforms
AT akhtaruzzamanm analyzingtheperformanceofdeeplearningmodelsfordetectinghatespeechonsocialmediaplatforms