Learning Semantic Network Patterns for Hypernymy Extraction

Current approaches of hypernymy acquisition are mostly based on syntactic or surface representations and extract hypernymy relations between surface word forms and not word readings. In this paper we present a purely semantic approach for hypernymy extraction based on semantic networks (SNs). This approach employs a set of patterns sub0(a1, a2) ← premise where the premise part of a pattern is given by a SN. Furthermore this paper describes how the patterns can be derived by relational statistical learning following the Minimum Description Length principle (MDL). The evaluation demonstrates the usefulness of the learned patterns and also of the entire hypernymy extraction system.