Legal W2V: Legal Vocabulary Embedding
收藏NIAID Data Ecosystem2026-03-12 收录
下载链接：
https://data.mendeley.com/datasets/97tytghd47
下载链接
链接失效反馈官方服务：
资源简介：
Legal W2V is a Legal domain-specific word embedding trained using more than 48,000 judgments from the Supreme Court of India, decided during the span period of January 1950 to December 2016 (76 years). It contains more than 34,000 Legal vocabulary words and their respective 100-dimensional vectors, which is trained using Continuous Bag of Word variant of Word2Vec, considering Gensim Library. Potential applications scope of this Legal Word embedding would be to explore various active research domains viz. Legal Information retrieval and Legal Informatics.
Parameter Settings: Vector size: 100, Window size: 10, Number of iterations (iter): 10, Minimum count (min_count): 10.
创建时间：
2021-05-28
