Igevorse at SemEval-2018 Task 10: Exploring an Impact of Word Embeddings Concatenation for Capturing Discriminative Attributes

Abstract

Semantic differences extraction is a challenging problem in Natural Language Processing and its solution is necessary for a realistic semantic representation as similarity information is not sufficient to capture individual aspects of meaning. This paper presents a comparison of several approaches for capturing discriminative attributes and considers an impact of concatenation of several word embeddings of different nature on the classification performance. A similarity-based method is proposed and compared with machine learning approaches. It is shown that this method outperforms others on all the considered word vector models and there is a performance increase when concatenated datasets are used.