Learning Class Disjointness Axioms Using Grammatical Evolution

Abstract

Today, with the development of the Semantic Web, Linked Open Data (LOD), expressed using the Resource Description Framework (RDF), has reached the status of "big data" and can be considered as a giant data resource from which knowledge can be discovered. The process of learning knowledge defined in terms of OWL 2 axioms from the RDF datasets can be viewed as a special case of knowledge discovery from data or "data mining", which can be called "RDF mining". The approaches to automated generation of the axioms from recorded RDF facts on the Web may be regarded as a case of inductive reasoning and ontology learning. The instances, represented by RDF triples, play the role of specific observations, from which axioms can be extracted by generalization. Based on the insight that discovering new knowledge is essentially an evolutionary process, whereby hypotheses are generated by some heuristic mechanism and then tested against the available evidence, so that only the best hypotheses survive, we propose the use of Grammatical Evolution, one type of evolutionary algorithm, for mining disjointness OWL 2 axioms from an RDF data repository such as DBpedia. For the evaluation of candidate axioms against the DBpedia dataset, we adopt an approach based on possibility theory.

An Evolutionary Approach to Class Disjointness Axiom Discovery

“…Here, we extend and enhance our recent proposal [14] of an approach to generate class disjointness axioms from an existing RDF repository using Grammatical Evolution (GE). On the one hand, the enhancement concerns the fitness function used to score axioms, where we now include an improved measure of generality and we removed the necessity measure, which, as we will explain below, does not carry any useful information when dealing with this type of axioms.…”

“…This incompleteness and noise determines a sort of epistemic uncertainty in the evaluation of the quality of a candidate axiom. In order to properly capture this type of uncertainty, typical of an open world, which contrasts with the ontic uncertainty typical of random processes, we follow [14] in adopting an axiom scoring heuristics based on possibility theory (cf. Section 3.3), which is well-suited to incomplete knowledge.…”

“…This appears to be a justified choice for assessing knowledge extracted from an RDF repository. We now summarize this scoring heuristics which was applied in [14] and we point out how our definition of the fitness function differs from [14].…”

“…In addition, what we are looking for is not only credible axioms, but also general ones. As suggested in [14], the generality of an axiom may be defined as the cardinality of its support, . In other words, an axiom is more general as the extension of its components contains more facts.…”

“…which was adopted by [14]. However, it should be noticed that negation is not supported by the syntax of RDF.…”