Everyone in our lab is free to work on what they want to work on. Encouragement and constructive feed back is always available.
On the Relationship between Feature Models and Ontologies
Title | On the Relationship between Feature Models and Ontologies |
Publication Type | Thesis |
Year of Publication | 2006 |
Authors | Kim, C H P. |
Advisor | Czarnecki, K. |
Academic Department | Electrical and Computer Engineering |
Degree | MASc |
Number of Pages | 89 |
University | University of Waterloo |
City | Waterloo |
Abstract | Feature modeling is an increasingly popular domain modeling technique, particularly used in software product line development for managing commonality and variability. Ontology modeling is also an increasingly popular domain modeling technique, applicable to software engineering in general. An emerging paradigm called model-driven software product lines (MDSPL) proposes systematic modeling as the often sought middle ground between configuration and custom coding of software product lines. In MDSPL, both feature modeling and ontology modeling influence solution models through domain modeling, but enigmatic differences between the two techniques, for example, in modeling philosophy and descriptive power, make their relationship intriguing. This relationship is explored in three stages. Firstly, a new light is shed on the nature of feature models, which clarifies exactly what feature models are and how they may be different from ontologies. Secondly, it is proposed that feature models are views on ontologies, where the mapping is defined precisely through syntactic correspondence and semantics. Finally, two complementary domain modeling approaches, view projection and view integration, are proposed that promote agility and address the limitations of feature modeling and ontology modeling. |
URL | http://www.cs.utexas.edu/~chpkim/chpkim-masc-thesis.pdf |