Analysing the Problem and Main Approaches for Ontology Population

Abstract

Knowledge systems are a suitable computational approach to solve complex problems and to provide decision support. Ontologies are an approach for knowledge representation and Ontology Population looks for instantiating the constituent elements of an ontology, like properties and non-taxonomic relationships. Manual population by domain experts and knowledge engineers is an expensive and time consuming task. Thus, automatic or semi-automatic approaches are needed. This paper discusses the problem of Automatic Ontology Population and proposes a generic process specifying its phases and what kind of techniques can be used to perform the activities of each phase. Some techniques representing the state of the art of this field are also described along with the solutions they adopt for each phase of the AOP process with their advantages and limitations. This work is part of HERMES, a Brazil/Portugal research cooperation project looking for techniques and tools for automating the process of ontology learning and population.

Investigating a Method for Automatic Construction and Population of Ontologies for Services: Performances and Limitations

“…Besides ontology construction, ontology population is another important aspect for SWS. Ontology population encompasses several phases, among which the candidate instances for population are identified, then classified inside the ontology [21]. Techniques used to accomplish those phases are similar to those used for ontology construction (NLP, IE, and some machine learning when a learning set is available).…”

Organization of Meta-knowledge in the Form of 23-Bit Templates for Big Data Processing

“…This organization for the Meta-knowledge of 23-bit templates adheres to the characteristics of Ontology Engineering in its ability to provide conceptualization of this methodological approach, provide explicitness in its parameters/boundaries, formalize the model for easy accessibility, while being comprehensible for the consensus of our field of study [8]. A domain specific, 23-bit meta-knowledge template provides us a clustering technique capable of processing large amounts of data.…”