Chinese Sexual-Related and Safe Text Dataset (10k Samples )

This dataset contains 10,000 Chinese text samples for research on the detection of sexual-content-related and safe text. It is built from 500 keywords associated with sexual-related semantics. For each keyword, 10 harmful samples contain explicit sexual-related terms, and 10 safe samples contain the same keyword in a non-sexual, contextually safe manner. The dataset is suitable for offensive language detection, sexual content recognition, text classification, and content moderation tasks in Chinese natural language processing.