Google News dataset
收藏arXiv2025-09-30 收录
下载链接：
https://code.google.com/archive/p/word2vec/
下载链接
链接失效反馈官方服务：
资源简介：
该数据集是一个用于提取名词对之间语义关系的语料库，它包含了300万词汇和短语的三维向量。此外，该数据集已经用于预训练Word2Vec向量，并且与SemEval-2010任务8的语义关系提取相匹配。它的任务是进行名词对之间多类别分类的语义关系分析。
This dataset is a corpus designed for extracting semantic relations between noun pairs. It contains 3-dimensional vectors for 3 million vocabulary terms and phrases. Additionally, this dataset has been utilized for pre-training Word2Vec vectors and aligns with the semantic relation extraction task of SemEval-2010 Task 8. Its core task is to conduct multi-class classification-based semantic relation analysis between noun pairs.
提供机构：
Google
