UWB at SemEval-2018 Task 10: Capturing Discriminative Attributes from Word Distributions

Proceedings of the 12th International Workshop on Semantic Evaluation (SemEval-2018), pages 935–939 New Orleans, Louisiana, June 5–6, 2018. ©2018 Association for Computational Linguistics UWB at SemEval-2018 Task 10: Capturing Discriminative Attributes from Word Distributions Tom´aˇs Brychc´ın1, Tom´aˇs Hercig1,2, Josef Steinberger2, and Michal Konkol1 1NTIS – New Technologies for the Information Society, Faculty of Applied Sciences, University of West Bohemia, Czech Republic 2Department of Computer Science and Engineering, Faculty of Applied Sciences, University of West Bohemia, Czech Republic {brychcin,tigi,jstein,konkol}@kiv.zcu.cz http://nlp.kiv.zcu.cz Abstract We present our UWB system for the task of capturing discriminative attributes at SemEval 2018. Given two words and an attribute, the system decides, whether this attribute is dis- criminative between the words or not. As- suming Distributional Hypothesis, i.e., a word meaning is related to the distribution across contexts, we introduce several approaches to compare word contextual information. We experiment with state-of-the-art semantic spaces and with simple co-occurrence statis- tics. We show the word distribution in the corpus has potential for detecting discrimina- tive attributes. Our system achieves F1 score 72.1% and is ranked #4 among 26 submitted systems. 1 Introduction In this paper, we describe our UWB system par- ticipating in the pilot shared task on capturing discriminative attributes held at SemEval 2018. Given two words and an attribute, the goal of this task is to decide, whether the attribute is discrim- inative between them. For example, we can dis- tinguish between the words car and boat by a dis- criminative feature (attribute) wheels. On the other hand, both tennis and basketball use a ball, so that the ball is not discriminative between them. By its nature, capturing discriminative attributes is a binary classification task. In general, there is no assumption on the input words and their attributes (e.g., part of speech, etc.). While most related works focus on extracting discriminative features from images (Guo et al., 2015; Huang et al., 2016; Lazaridou et al., 2016), this shared task is oriented purely on textual level. The first experiments have been performed by Krebs and Paperno (2016) and have shown the promising potential of this task. The fundamental assumption of our work is Distributional Hypothesis, i.e., two words are ex- pected to be semantically similar if they occur in similar contexts (they are similarly distributed across the text). This hypothesis was formulated by Harris (1954) several decades ago. Today it is the basis of state-of-the-art distributional semantic models (Mikolov et al., 2013; Pennington et al., 2014; Bojanowski et al., 2017). We present several approaches, which rely on Distributional Hypoth- esis and employ the word contexts for statistical comparison of their meanings. 2 Proposed Approach Given two words w1 ∈ V , w2 ∈ V and the at- tribute a ∈ V , where V is a word vocabulary. The task is to predict, whether the attribute a is dis- criminative between the words w1 and w2, which leads to a binary classification task. We propose several metrics, which estimate the degree to which the attribute a is important for …

Content truncated to 500 words