Improving Hypernymy Detection with an Integrated Path-based and Distributional Method

Abstract

Detecting hypernymy relations is a key task in NLP, which is addressed in the literature using two complementary approaches. Distributional methods, whose supervised variants are the current best performers, and path-based methods, which received less research attention. We suggest an improved path-based algorithm, in which the dependency paths are encoded using a recurrent neural network, that achieves results comparable to distributional methods. We then extend the approach to integrate both pathbased and distributional signals, significantly improving upon the state-of-the-art on this task.

SemEval-2018 Task 9: Hypernym Discovery

“…1 Most of the recent work on the subject is however supervised, and in the main based on using word embeddings as input for classification or prediction (e.g Baroni et al, 2012;Santus et al, 2014;Fu et al, 2014;Weeds et al, 2014;Sanchez Carmona and Riedel, 2017;Nguyen et al, 2017). As shown by Shwartz et al (2016), pattern-based and distributional evidences can be effectively combined within a neural architecture. In this shared task we have actually received systems of both natures, including a combination of pattern-based and distributional cues, similar to the one mentioned above, which also proved to be highly effective (see Section 5).…”

“…the capability to relate generic terms or classes to their specific instances, lies at the core of human cognition. It is not surprising, therefore, that identifying hypernymic (is-a) relations has been pursued in NLP for more than two decades (Shwartz et al, 2016): indeed, successfully identifying this lexical relation substantially improves Question Answering applications (Prager et al, 2008;Yahya et al, 2013), Textual Entailment and Semantic Search systems (Hoffart et al, 2014;Roller et al, 2014;Roller and Erk, 2016). In addition, hypernymic relations are the backbone of almost every ontology, semantic network and taxonomy (Yu et al, 2015), which are in turn useful resources for downstream tasks such as web retrieval, website navigation or records management (Bordea et al, 2015).…”

Exploring Vector Spaces for Semantic Relations

“…These works either aim to identify very specific relation types (typically taxonomical relations) with a mixture of features and a supervised classifier, or target analogy pairs: a task in which, as we have seen, relation-unaware baselines approximate relation-aware representations. However, more recently, Shwartz et al (2016) achieved promising results on the hypernymy detection task by combining dependency path-based context representations with distributional vectors; this finding can be relevant for a broader range of semantic relations as well.…”

Classifying Lexical-semantic Relationships by Exploiting Sense/Concept Representations

“…A number of research efforts have been devoted to developing an automated and accurate method to type the relationship between an arbitrary pair of words. Most of these studies (Fu et al, 2014;Kiela et al, 2015;Shwartz et al, 2016), however, concentrated on the hypernymy relation, since it is the most fundamental relationship that forms the core taxonomic structure in a lexical-semantic resource. In comparison, fewer studies considered a broader range of lexical-semantic relations, e.g., (Necsuleşcu et al, 2015) and our present work.…”