Ontological analysis for dynamic data model exploration


Increasing access to data and computational resources allows use to use more expressive approaches to data analysis. We propose using established statistical metrics to assist the automatic analysis of free text transcripts. The meaningful concepts from a domain and their axiomatic relationships can be captured in an ontology. This provides an aggregate model which describes the domain. However, the fine detail from individual elements and their characteristics are subsumed by the whole.  Keeping multiple 'micro models' of the data, along with meta information allows a range of different view points. This can be applied to free text documents that within a domain where significant information is carried by one or a few instances such as in the analysis of interview transcripts. This paper presents a framework that utilises ontological tools to create domain models in a way that it allows for a distributed and parallel implementation necessary for big data analysis.

DOI Code: 10.1285/i2037-3627v5n1p42

Keywords: Ontology, model, classifier, concept mapping

Full Text: pdf

Creative Commons License
This work is licensed under a Creative Commons Attribuzione - Non commerciale - Non opere derivate 3.0 Italia License.