文档介绍：Ontology-Based Integration of Information A Survey of Existing Approaches
, , , , , and
University of Bremen
Motivation
Vast information is available on the WWW
Growing need for
Finding relevant information (Information Extraction)
Creating new knowledge out of the available information (Web Content Mining)
Personalization of the web
Learning about customers or individual users (Web Usage Mining)
Issues
Information is widely distributed and heterogeneous
Schema discovery
Wrapping
anizing data sources
Coping with changes in sources
Issues (cont.)
Structural (schematic) heterogeneity
Data is stored in different structures across the information systems
Semantic (data) heterogeneity
Considers the content of an information and its intended meaning
Causes
Confounding conflicts
Items that have the same meaning but differ in reality
Scaling conflicts
Different reference systems are used to measure a value (Eu, $)
Naming conflicts
Naming schemes of similar items differ significantly
Solution - Ontologies
Refers to shared understanding of a domain of interest which may be used as a unifying framework
Embodies some sort of world view with respect to a given domain
World view is conceived as:
Set of concepts (entities, attributes, processes)
Definitions
Inter-relationships
This is referred to as conceptualization
Ontologies (cont.)
consensual, shared and formal description of the concepts that are important in a given domain
identifies classes of objects that are important in a domain anizes these classes in a subclass hierarchy
each class is characterized by properties shared by all elements in that class
important relations between classes or between the elements of the classes are also part of an ontology
Ontology – example
Objective
Evaluate the use of ontologies in information integration systems
SIMS, TSIMMIS, OBXERVER, CARNOT, Infosleuth, KRAFT, PICSEL, DWQ, Ontobroker, SHOE