Neural Vector Conceptualization for Word Vector Space Interpretation

Robert Schwarzenberg, Lisa Raithel, David Harbecke

In: NAACL-HLT 2019 Workshop on Evaluating Vector Space Representations for NLP (RepEval). Meeting of the North American Chapter of the Association for Computational Linguistics (NAACL) June 2-7 MInneapolis Minnesota United States Association for Computational Linguistics 2019.


Distributed word vector spaces are considered hard to interpret which hinders the under-standing of natural language processing (NLP) models. In this work, we introduce a new method to interpret arbitrary samples from a word vector space. To this end, we train a neural model to conceptualize word vectors, which means that it activates higher order concepts it recognizes in a given vector. Contrary to prior approaches, our model operates in the original vector space and is capable of learning non-linear relations between word vectors and concepts. Furthermore, we show that it produces considerably less entropic concept activation profiles than the popular cosine similarity.


neural_vector_conceptualization.pdf (pdf, 577 KB)

German Research Center for Artificial Intelligence
Deutsches Forschungszentrum für Künstliche Intelligenz