Ontology-Based Feature Selection: A Survey

The Semantic Web emerged as an extension to the traditional Web, adding meaning (semantics) a distributed of structured and linked information. At its core, concept ontology provides means semantically describe structure information, expose it software human agents in machine human-readable form. For be realized, is crucial develop powerful artificial intelligence machine-learning techniques, able extract knowledge from information sources, represent underlying ontology. This survey aims provide insight into key aspects ontology-based extraction various sources such text, databases, expertise, realized realm feature selection. First, common classification selection algorithms are presented. Then, selected approaches, which utilize ontologies features perform classification, described. selective representative approaches span diverse application domains, document opinion mining, manufacturing, recommendation systems, urban management, security demonstrate feasibility applicability methods. survey, addition criteria-based presentation related works, contributes number open issues challenges this still active research topic.