Scholarlink

As digitization continues to proliferate globally, individuals prefer expressing themselves on social media platforms using their native languages. On these platforms, people share their interests, thoughts, and rights, sometimes receiving appreciation for their views and, at times, encountering conflicts of interest leading to mean comments, including hate speech and aggression towards individuals, societies, or groups. Detecting such comments has become crucial to curbing further abuse. This research paper focuses on identifying aggressive and non-aggressive Bengali text in social media posts and comments through the application of machine learning and deep learning algorithms. The dataset was collected from various social media platforms like Facebook, YouTube, and Twitter. Employing diverse machine learning and deep learning algorithms, such as SVM, Random Forest, KNN, Linear Regression, Decision Tree, and CNN, the authors achieved the highest accuracy of 90.46% with Multinomial Naive Bayes (MNB).

错误：搜索内容不能为空，请输入英文关键词
错误：关键词超出字数限制，请精简
高级检索

Aggressive Bangla Text Detection Using Machine Learning and Deep Learning Algorithms

  • 会议论文

  • 发表日期：

    2025年

查看全文
  • Tanjela Rahman Rosni,
  • Mahamudul Hasan,
  • Tanni Mittra,
  • Md. Sawkat Ali,
  • Md. Hasanul Ferdaus

摘要

As digitization continues to proliferate globally, individuals prefer expressing themselves on social media platforms using their native languages. On these platforms, people share their interests, thoughts, and rights, sometimes receiving appreciation for their views and, at times, encountering conflicts of interest leading to mean comments, including hate speech and aggression towards individuals, societies, or groups. Detecting such comments has become crucial to curbing further abuse. This research paper focuses on identifying aggressive and non-aggressive Bengali text in social media posts and comments through the application of machine learning and deep learning algorithms. The dataset was collected from various social media platforms like Facebook, YouTube, and Twitter. Employing diverse machine learning and deep learning algorithms, such as SVM, Random Forest, KNN, Linear Regression, Decision Tree, and CNN, the authors achieved the highest accuracy of 90.46% with Multinomial Naive Bayes (MNB).