SpectralWords: Spectral Embeddings Approach to Word Similarity Task for Large Vocabularies

Ivan Lobov

Abstract: In this paper we show how recent advances in spectral clustering using Bethe Hessian operator can be used to learn dense word representations. We propose an algorithm SpectralWords that achieves comparable to the state-of-the-art performance on word similarity tasks for medium-size vocabularies and can be superior for datasets with larger vocabularies.

Keywords: spectral clustering, distributed representation, embeddings, word similarities

TL;DR: Beating Skip-gram and SVD (on PPMI) on word similarity tasks with large vocabularies by using spectral-based approach.