Pavel Smrz, Brno University of Technology, Czech Republic

This paper discusses a new approach to creating semantic resources consisting of complex associations among words that can be used for evaluating the content of word embeddings as well as in various language-learning scenarios. We briefly introduce Codenames – an existing party board game – and the way of recording word associations suggested by human players. Advanced word embedding models are then compared on the collected data and it is demonstrated that they often fail in the cases of complex word associations that go beyond simple contextual interchangeability. We conclude with an initial evaluation of the automatic guessing of associated words based on clues provided by human players and a discussion on further extensions of the system towards a wide language coverage and explanations of word associations in the language learning context.

Natural Language Processing, Word Embedding, Distributional Semantics, Implicit Crowdsourcing, Games with Purpose, Semantic Representation