Non-distributional Word Vector Representations
This addresses the need for more transparent and linguistically grounded word representations in NLP, though it is incremental as it builds on existing resources and evaluation methods.
The paper tackles the problem of uninterpretable word vectors in NLP by constructing interpretable binary vectors from hand-crafted linguistic resources like WordNet and FrameNet, achieving competitive performance to standard distributional approaches on state-of-the-art evaluations.
Data-driven representation learning for words is a technique of central importance in NLP. While indisputably useful as a source of features in downstream tasks, such vectors tend to consist of uninterpretable components whose relationship to the categories of traditional lexical semantic theories is tenuous at best. We present a method for constructing interpretable word vectors from hand-crafted linguistic resources like WordNet, FrameNet etc. These vectors are binary (i.e, contain only 0 and 1) and are 99.9% sparse. We analyze their performance on state-of-the-art evaluation methods for distributional models of word vectors and find they are competitive to standard distributional approaches.