Global Vector

Year: 2,014
Journal: Conference on Empirical Methods in Natural Language Processing
Languages: All Languages
Programming languages: Shell
Input data:


Output data:

words as vectors

GloVe, coined from Global Vectors, is a model for distributed word representation. The model is an unsupervised learning algorithm for obtaining vector representations for words. This is achieved by mapping words into a meaningful space where the distance between words is related to semantic similarity. Training is performed on aggregated global word-word co-occurrence statistics from a corpus, and the resulting representations showcase interesting linear substructures of the word vector space. It is developed as an open-source project at Stanford. As log-bilinear regression model for unsupervised learning of word representations, it combines the features of two model families, namely the global matrix factorization and local context window methods.

Sign In


Reset Password

Please enter your username or email address, you will receive a link to create a new password via email.