The ContrastMedium Algorithm: Taxonomy Induction From Noisy Knowledge Graphs With Just A Few Links

Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 590–600, Valencia, Spain, April 3-7, 2017. c⃝2017 Association for Computational Linguistics The ContrastMedium Algorithm: Taxonomy Induction From Noisy Knowledge Graphs With Just a Few Links Stefano Faralli1, Alexander Panchenko2, Chris Biemann2 and Simone Paolo Ponzetto1 1Data and Web Science Group, University of Mannheim, Germany 2Language Technology Group, University of Hamburg, Germany {stefano,simone}@informatik.uni-mannheim.de {panchenko,biemann}@informatik.uni-hamburg.de Abstract In this paper, we present ContrastMedium, an algorithm that transforms noisy se- mantic networks into full-fledged, clean taxonomies. ContrastMedium is able to identify the embedded taxonomy structure from a noisy knowledge graph without ex- plicit human supervision such as, for in- stance, a set of manually selected input root and leaf concepts. This is achieved by leveraging structural information from a companion reference taxonomy, to which the input knowledge graph is linked (either automatically or manually). When used in conjunction with methods for hypernym acquisition and knowledge base linking, our methodology provides a complete so- lution for end-to-end taxonomy induction. We conduct experiments using automati- cally acquired knowledge graphs, as well as a SemEval benchmark, and show that our method is able to achieve high perfor- mance on the task of taxonomy induction. 1 Introduction Recent years have witnessed an impressive amount of work on automatic construction of wide-coverage knowledge resources. Web-scale open information extraction systems like NELL (Carlson et al., 2010) or ReVerb (Fader et al., 2011) have been successful in acquiring massive amounts of machine-readable knowledge by ef- fectively tapping large amounts of text from Web pages. However, the output of these systems does not consist of a clean, fully-semantified output. Such output, on the other hand, could be pro- vided by the vocabulary of large-scale ontologies like DBpedia (Bizer et al., 2009) or YAGO (Hof- fart et al., 2013) and the integration of open and closed information extraction approaches (Dutta et al., 2014). The use of an encyclopedia-centric (e.g., Wikipedia-based) dictionary of entities leads to poor coverage of domain-specific terminologies (Faralli and Navigli, 2013). This can be allevi- ated by constructing knowledge bases of ever in- creasing coverage and complexity from the Web (Wu et al., 2012; Gupta et al., 2014; Dong et al., 2014) or by community efforts (Bollacker et al., 2008). However, the focus on large size and wide coverage at entity level has led all these resources to avoid the complementary problem of curat- ing and maintaining a clean taxonomic backbone with as minimal supervision as possible. That is, no resource, to date, integrates structured infor- mation from existing wide-coverage knowledge graphs with empirical evidence from text for the explicit goal of building full-fledged taxonomies consisting of a clean and fully-connected directed acyclic graph (DAG). This is despite the fact that taxonomies have been known for a long time to provide valid tools to represent domain-specific knowledge with dozens of scientific, industrial and social applications (Glass and Vessey, 1995). In taxonomy induction, the required domain knowledge can be acquired with …

Content truncated to 500 words