Determining Gains Acquired from Word Embedding Quantitatively using Discrete Distribution Clustering

Jianbo Ye (1), Yanran Li (2), Zhaohui Wu (1), James Z. Wang (1), Wenjie Li (2) and Jia Li (1)

(1)The Pennsylvania State University, USA
(2)The Hong Kong Polytechnic University, Hong Kong

Word embeddings have become widelyused in document analysis. While a large number of models for mapping words to vector spaces have been developed, it remains undetermined how much net gain can be achieved over traditional approaches based on bag-of-words. In this paper, we propose a new document clustering approach by combining any word embedding with a state-of-the-art algorithm for clustering empirical distributions. By using the Wasserstein distance between distributions, the word-to-word semantic relationship is taken into account in a principled way. The new clustering method is easy to use and consistently outperforms other methods on a variety of data sets. More importantly, the method provides an effective framework for determining when and how much word embeddings contribute to document analysis. Experimental results with multiple embedding models are reported.

Full Paper
(PDF, 0.4MB)

(ZIP, 15MB)

Source Codes for Scripts to Process Data

Citation: Jianbo Ye, Yanran Li, Zhaohui Wu, James Z. Wang, Wenjie Li and Jia Li, ``Determining Gains Acquired from Word Embedding Quantitatively using Discrete Distribution Clustering,'' Proceedings of the Annual Meeting of the Association for Computational Linguistics, vol. 1, pp. 1847-1856, Vancouver, Canada, August 2017.

© 2017 Association for Computational Linguistics (ACL). Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the ACL.

Last Modified: April 3, 2017
© 2017