Interactive Re-Fitting as a Technique for Improving Word Embeddings

Word embeddings are a fixed, distributional representation of the context of words in a corpus learned from word co-occurrences. While word embeddings have proven to have many practical uses in natural language processing tasks, they reflect the attributes of the corpus upon which they are trained. Recent work has demonstrated that post-processing of word embeddings to apply information found in lexical dictionaries can improve their quality. We build on this post-processing technique by making it interactive. Our approach makes it possible for humans to adjust portions of a word embedding space by moving sets of words closer to one another. One motivating use case for this capability is to enable users to identify and reduce the presence of bias in word embeddings. Our approach allows users to trigger selective post-processing as they interact with and assess potential bias in word embeddings.