A Survey on Ontology Creation Methodologies

Abstract

In the current literature of knowledge management and artificial intelligence, several different approaches to the problem have been carried out of developing domain ontologies from scratch. All these approaches deal fundamentally with three problems: (1) providing a collection of general terms describing classes and relations to be employed in the description of the domain itself; (2) organizing the terms into a taxonomy of the classes by the ISA relation; and (3) expressing in an explicit way the constraints that make the ISA pairs meaningful. Though a number of such approaches can be found, no systematic analysis of them exists which can be used to understand the inspiring motivation, the applicability context, and the structure of the approaches. In this paper, we provide a framework for analyzing the existing methodologies that compares them to a set of general criteria. In particular, we obtain a classification based upon the direction of ontology construction; bottom-up are those methodologies that start with some descriptions of the domain and obtain a classification, while top-down ones start with an abstract view of the domain itself, which is given a priori. The resulting classification is useful not only for theoretical purposes but also in the practice of deployment of ontologies in Information Systems, since it provides a framework for choosing the right methodology to be applied in the specific context, depending also on the needs of the application itself.

A Generic Model of Ontology to Visualize Information Science Domain (OIS)

“…The approach depends on the purpose and kind of ontology being developed. The evaluation process is necessary even when building an ontology for a particular domain from the beginning or modifying an existing one (Gòmez-Pérez et al, 2004, Cristani andCuel, 2005) .…”

Bottom-Up Extraction and Trust-Based Refinement of Ontology Metadata

“…More recently, several techniques specifically aimed at learning ontologies from text corpora or database repositories were proposed [5]. Well-known research approaches to ontology learning [6] include:…”

Mapping longitudinal studies to risk factors in an ontology for dementia

“…While there are different approaches to constructing ontologies [5], a common approach is to identify the basic (dementia) concepts, describe the properties of these concepts, and generate a vocabulary of concept-related keywords. This process here is referred to as Ontology Initialization, see Figure 1, and the terms in each stage of initialisation map directly to those used in each of our three stages of matching.…”