WebBase

该数据集是一个大规模的英语语料库，包含了大约30亿个标记，被用于评估所提出的WEQ方法。在预处理后，词汇量达到了277,704个单词。该数据集的规模为30亿个标记，适用于多种任务，包括词语相似度计算、文本分类和命名实体识别。

This dataset is a large-scale English corpus containing approximately 3 billion tokens, which is used to evaluate the proposed WEQ method. After preprocessing, its vocabulary size reaches 277,704 words. Boasting a scale of 3 billion tokens, this dataset supports a variety of tasks including word similarity calculation, text classification and named entity recognition.