Neural Word Embedding as Implicit Matrix Factorization

Explore this paper's citation graph

Summary

It is shown that using a sparse Shifted Positive PMI word-context matrix to represent words improves results on two word similarity tasks and one of two analogy tasks, and conjecture that this stems from the weighted nature of SGNS's factorization.

Type
article
Published
2014-12-08
Cited by
2,049
References
30

Keywords

Word (group theory), Word embedding, Computer science, Context (archaeology), Matrix decomposition

References

Cited by

Related papers