Distributional semantics for ontology verification

Proceedings of the Fourth Joint Conference on Lexical and Computational Semantics (*SEM 2015), pages 30–39, Denver, Colorado, June 4–5, 2015. Distributional semantics for ontology verification ∗ Julien Corman IRIT, University of Toulouse julien.corman@irit.fr Nathalie Aussenac-Gilles CNRS IRIT, University of Toulouse aussenac@irit.fr Laure Vieu CNRS IRIT, University of Toulouse vieu@irit.fr Abstract As they grow in size, OWL ontologies tend to comprise intuitively incompatible statements, even when they remain logically consistent. This is true in particular of lightweight on- tologies, especially the ones which aggregate knowledge from different sources. The article investigates how distributional semantics can help detect and repair violation of common sense in consistent ontologies, based on the identification of consequences which are un- likely to hold if the rest of the ontology does. A score evaluating the plausibility for a con- sequence to hold with regard to distributional evidence is defined, as well as several methods in order to decide which statements should be preferably amended or discarded. A conclu- sive evaluation is also provided, which con- sists in extending an input ontology with ran- domly generated statements, before trying to discard them automatically. 1 Introduction Ontology learning from texts deals with the auto- mated extraction of knowledge from linguistic ev- idence. This article investigates a slightly differ- ent problem, which is how Natural Language Pro- cessing may provide hints for the identification of statements of an input ontology which are unlikely to hold if the rest of it does. As a minimal exam- ple, consider the following set ∆ of statements, from DBpedia (Mendes et al., 2012), and assume that ∆ is ∗ The research reported here was supported by a Marie Curie FP7 Career Integration Grant, Grant Agreement Number PCIG13-GA-2013-618550. a subset of a larger set of statements K (for instance DBpedia itself, or some subset of it) : Ex 1. ∆ = {(1) keyPerson(Caixa Bank, CEO), (2) keyPerson(BrookField Office Properties, Peter Munk) (3) occupation(Peter Munk, CEO) } There is a clear violation of common sense in ∆ : the individual CEO must be both a key person of Caixa Bank, and the occupation of another individ- ual (Peter Munk), who is himself a key person of some company. Detecting such cases within (larger) sets of logical statements is of particular interest in OWL, which facilitates the aggregation of knowl- edge from multiple sources with overlapping signa- tures, yielding datasets in which several incompati- ble understandings of a same individual or predicate may coexist. This easily leads to undesired infer- ences, even when the dataset is logically consistent.1 But as the example illustrates, the problem may also occur within a single knowledge base, especially if it has been built semi-automatically, and/or is issued from a collaborative effort. Another problem of interest consists in deciding which statement(s) should be preferably discarded or amended in order to get rid of the nonsense. In example 1, without further information, it would be intuitively relevant to discard or modify either (1) or (2). Unfortunately though, ∆ alone does not give any …

Content truncated to 500 words