Dataset for Emotionally Manipulative vs. Non-Manipulative Chinese Texts

Dataset for Emotionally Manipulative vs. Non-Manipulative Chinese Texts 【This dataset was originally released on Hugging Face and archived in Zenodo to provide a permanent DOI reference】 Summary This dataset consists of 10,000 Chinese texts evenly split into 5,000 manipulative and 5,000 non-manipulative samples. It was constructed to support research on detecting emotionally manipulative language. Data Fields id: unique identifier text: Chinese sentence label: manipulative(1) or non-manipulative(0) Intended Use For academic research and educational purposes, with a focus on NLP tasks such as emotion, persuasion, and manipulation detection. Citation Please cite this Zenodo record when using the dataset. The dataset is made available for academic and non-commercial use.