Encoding Prior Knowledge with Eigenword Embeddings
Published
2016-07-29
Dominique Osborne
,
Shashi Narayan
,
Shay B. Cohen
Dominique Osborne
University of Strathclyde
Shashi Narayan
University of Edinburgh
Shay B. Cohen
University of Edinburgh
Abstract
Canonical correlation analysis (CCA) is a method for reducing the dimension of data represented using two views. It has been previously used to derive word embeddings, where one view indicates a word, and the other view indicates its context. We describe a way to incorporate prior knowledge into CCA, give a theoretical justification for it, and test it by deriving word embeddings and evaluating them on a myriad of datasets.