World Central Banks (WCB) dataset

该数据集名为WCB，是迄今为止最全面的政策性货币语料库，包含了来自全球不同地理区域的25家中央银行的超过38万句句子，时间跨度长达28年。该数据集采用了双重标注者进行标注，经过了广泛的审查流程，并包含了详尽的标注指南框架。规模上，它汇集了来自25家中央银行的超过38万句句子。该数据集的任务包括立场检测、时间分类和不确定性估计。This dataset, named WCB, is the most comprehensive monetary policy corpus to date. It contains over 380,000 sentences from 25 central banks across diverse global regions, with a time span of 28 years. The dataset uses dual annotators for annotation, has undergone extensive review processes, and includes a comprehensive annotation guideline framework. In terms of scale, it compiles over 380,000 sentences from 25 central banks. The tasks supported by this dataset include stance detection, temporal classification, and uncertainty estimation.