Access Type

Open Access Dissertation

Date of Award

January 2015

Degree Type


Degree Name



Computer Science

First Advisor

Zaki Malik


This thesis focuses on (1) accessing relational databases through Semantic Web technologies and (2) resolving conflicts that usually arises when integrating data from heterogeneous source schemas and/or instances.

In the first part of the thesis, we present an approach to access relational databases using Semantic Web technologies. Our approach is built on top of Ontop framework for Ontology Based Data Access. It extracts both Ontop mappings and an equivalent OWL ontology from an existing database schema. The end users can then access the underlying data source through SPARQL queries. The proposed approach takes into consideration the different relationships between the entities of the database schema when it extracts the mapping and the equivalent ontology. Instead of extracting a flat ontology that is an exact copy of the database schema, it extracts a rich ontology. The extracted ontology can also be used as an intermediary between a domain ontology and the underlying database schema. Our approach covers independent or master entities that do not have foreign references, dependent or detailed entities that have some foreign keys that reference other entities, recursive entities that contain some self references, binary join entities that relate two entities together, and n-ary join entities that map two or more entities in an n-ary relation. The implementation results indicate that the extracted Ontop mappings and ontology are accurate. i.e., end users can query all data (using SPARQL) from the underlying database source in the same way as if they have written SQL queries.

In the second part, we present an overview of the conflict resolution approaches in both conventional data integration systems and collaborative data sharing communities. We focus on the latter as it supports the needs of scientific communities for data sharing and collaboration. We first introduce the purpose of the study, and present a brief overview of data integration. Next, we talk about the problem of inconsistent data in conventional integration systems, and we summarize the conflict handling strategies used to handle such inconsistent data. Then we focus on the problem of conflict resolution in collaborative data sharing communities. A collaborative data sharing community is a group of users who agree to share a common database instance, such that all users have access to the shared instance and they can add to, update, and extend this shared instance. We discuss related works that adopt different conflict resolution strategies in the area of collaborative data sharing, and we provide a comparison between them. We find that a Collaborative Data Sharing System (CDSS) can best support the needs of certain communities such as scientific communities. We then discuss some open research opportunities to improve the efficiency and performance of the CDSS. Finally, we summarize our work so far towards achieving these open research directions.