Data Governance

93

5

Semantic Analytics and Ontologies

O. Takaki, N. Izumi, K. Murata, and K. Hasida

CONTENTS

Introduction ......................................................................................................94

Ontology ............................................................................................................95

Denition of Ontologies ............................................................................ 96

Purposes for Constructing Ontologies .................................................... 96

Construction of Ontologies ........................................................................98

Concepts (Classes) ..................................................................................98

Properties (Slots) .................................................................................... 99

Axioms .................................................................................................. 100

Representation of Linguistic Vocabulary on a Semantic Web ............ 100

RDF (Resource Description Framework) .......................................... 101

RDFS (Resource Description Framework Schema) .........................101

OWL (Web Ontology Language) ........................................................103

Semantic Analytics .........................................................................................104

Semantic Analysis Based on Natural Language Text ............................106

Semantic Analysis Based on ConditionalExpression of

Graph-Based Representation ...................................................................107

Semantic Analytics on a Semantic Web ..................................................108

A Framework for Denition and Calculation of Quality Indications .....108

Quality Indicators ......................................................................................109

Overview of QI-Framework ..................................................................... 110

Outline of QI-RS ........................................................................................111

Medical Service Ontology ........................................................................112

Outline of MSO Concepts ...................................................................112

Patients ...................................................................................................112

Events .....................................................................................................114

States of Patients ....................................................................................115

Main Relations in MSO ........................................................................115

94  •  O. Takaki, N. Izumi, K. Murata, and K. Hasida

INTRODUCTION

is chapter explains semantic analytics in data governance (DG) by

introducing a framework to dene quality indicators and to calculate

their values based on medical databases, where quality indicators are

measures of medical service quality, which are represented by numerical

values . Most importantly, we introduce an ontology called Medical

Service Ontology (MSO) as an example of an ontology that plays the

central role in semantic analytics.

Semantic analytics plays an important role in DG. e term semantic

analytics in this chapter refers to a technique used for semantically

analyzing, retrieving, integrating, or managing data resources in several

databases and on the Internet using ontologies. In fact, it is one of DG’s

primary roles to manage and utilize the data accumulated by an organiza-

tion and to use that data for the organizational decision making. However,

for this purpose, it is essential to be able to deal with data in an integrated

manner beyond dierences in data formats or expressions. Semantic

analytics judges the semantic identity or similarity between data beyond

syntactic dierences, making it possible to collectively deal with the same

or similar data from data resources in various formats. Moreover, ontolo-

gies are important as the fundamental tools of current semantic analysis.

We here explain a role of ontology in semantic analytics by natural lan-

guage processing (NLP). NLP is an area of research and application that

explores how computers can be used to understand and manipulate natu-

ral language text or speech to do useful things [Chowdhury, 2003]. NLP

can be regarded as a basic theory of semantic analytics. Knowledge used

in the four stages of analysis in NLP—morphological analysis, syntactic

analysis, semantic analysis, and context analysis—can roughly be divided

Objective Graphs and Quantifying Concepts ........................................ 117

Objective Graphs ..................................................................................117

Quantifying Concepts ..........................................................................117

Example of a Quality Indicator in QI-RS ..........................................117

Related Work .................................................................................................. 120

Conclusions .....................................................................................................121

Acknowledgement .......................................................................................... 121

References ........................................................................................................ 121