Aggression-annotated Corpus of Hindi-English Code-mixed Data

本数据集名为‘Aggression-annotated Corpus of Hindi-English Code-mixed Data’，由安贝德卡尔大学和印度信息技术学院-斯里城等机构合作创建。数据集主要收集自印度最流行的社交媒体平台Twitter和Facebook，包含约18k条推文和21k条Facebook评论，涉及多种语言，包括英语、印地语及混合语言。创建过程中，数据集经过精心标注，使用三级标签系统，详细区分不同类型的攻击行为。该数据集旨在帮助研究网络攻击行为，如网络欺凌、仇恨言论等，以期通过自动识别技术减少此类行为的发生。

This dataset is named 'Aggression-annotated Corpus of Hindi-English Code-mixed Data', and was co-developed by Ambedkar University, Indian Institute of Information Technology Sri City, and other collaborating institutions. The dataset is primarily collected from Twitter and Facebook, two of the most popular social media platforms in India, and contains approximately 18,000 tweets and 21,000 Facebook comments, covering multiple language varieties including English, Hindi, and code-mixed texts. During the creation process, the dataset was meticulously annotated using a three-level labeling system to thoroughly distinguish between different types of aggressive behaviors. This dataset aims to support research on cyber aggression-related issues such as cyberbullying and hate speech, with the ultimate goal of reducing the occurrence of such harmful behaviors through automated recognition technologies.