Multi-Label Extremism and Jihadism Classification Tweets Dataset

The "Multi-Label Extremism and Jihadism Classification Tweets Dataset" dataset is a multilingual resource designed for multi-label classification of online extremism and toxic behavior, including extremism and jihadism. Each comment is annotated with labels indicating the presence of various extremism traits: toxic, severe toxic, obscenity, threats, insults, identity hate, and jihadi content. This dataset is valuable for research in automated content moderation, enabling the detection of harmful and extremist content across multiple languages, and contributing to the development of safer online environments by providing a diverse array of real-world examples.