kubraElahi/Contribution-

Bangla Religious Aggressive Comments Dataset # 🧠 Bangla Religious Aggressive Comments Dataset (BRAC) This dataset contains Bangla-language comments labeled for aggression and targeted religious hate Comments. It is designed to support research in Natural Language Processing (NLP), including aggression detection, hate speech moderation, and classification tasks focused on religious tolerance in online spaces. # 📂 Dataset Overview Language: Bangla (বাংলা) Domain: Social Media, Online Comments Total Samples: [add total rows, e.g. ~2,000] Purpose: Aggression classification and religious hate speech analysis # 🧾 Data Format Each row in the dataset includes: Column Description Comments The Bangla-language comment text Aggression Label One of: Aggressive, Non-Aggressive Target Religion The religion targeted in the comment (e.g., Hindu, Muslim) # 🔍 Sample Data Comments,Aggression Label,Target Religion তোরা কাফের, বাংলাদেশে থাকতে পারবি না।,Aggressive,Hindu এটা সবার ধর্মের প্রতি সম্মান জানানোর দিন।,Non-Aggressive, ধর্ম নিয়ে কথা বলবি তোকে কুত্তার মতো মারব।,Aggressive,Islam # 🎯 Use Cases This dataset is useful for: Hate speech and aggression detection in Bangla Religious bias classification Social media moderation tools Multilingual or cross-lingual NLP training Transfer learning for low-resource languages languages # 🧑‍💻 How to Use You can load and analyze the dataset using pandas: import pandas as pd df = pd.read_csv("Bangla_religious_aggressive_comments.csv") print(df.head()) # 📚 Annotation Guidelines [-] Aggressive: Includes threats, hate, or abuse. Non-Aggressive: Respectful, neutral, or irrelevant comments. Target Religion: Left blank if no religion is targeted. Each comment was manually reviewed and annotated by native Bangla speakers. # 📈 Statistics (Fill these out based on your data) Aggressive: 10,000 Non-Aggressive: 10,000 Targeted Religions: Hindu: 2500 Islam: 2500 Christian: 2500 Buddhist: 2500 # 🪪 License This dataset is released under the Creative Commons Attributio …