QCRI/CrisisBench-english

CrisisBench数据集是一个用于灾害信息处理的基准数据集，特别关注社交媒体数据的文本分类。该数据集整合了多个来源的数据，包括CrisisLex、CrisisNLP等，并进行了类标签映射和去重处理。数据集包含英文语言的数据，格式为JSON，每个JSON对象包含id、event、source、text、lang、lang_conf和class_label等字段。数据集的目的是为社区提供基准结果，并支持灾害响应和危机管理的研究。

The CrisisBench dataset is a benchmark dataset for humanitarian information processing, particularly focusing on text classification of social media data. It integrates data from multiple sources, including CrisisLex, CrisisNLP, and others, and has undergone class label mapping and deduplication. The dataset contains English language data in JSON format, with each JSON object including fields such as id, event, source, text, lang, lang_conf, and class_label. The purpose of the dataset is to provide benchmark results for the community and support research in disaster response and crisis management.