Bachelor and Master Theses

Title: Master thesis: Measuring ontology awareness of model differencing approaches.
Subject: Computer Science
Level: Advanced
Description: Model-Driven Engineering promotes models as main artefacts of software development. In this respect, models are expected to evolve, analogously to code, whenever the application required to be maintained. Given the abstraction level of models, comparing two models in order to derive the evolution that caused the old version to be changed towards the new one is not always straightforward. In particular, if differences have been produced in a distributed environment, a proper differencing algorithm should be able to match corresponding elements in the two subsequent versions and generate an appropriate sequence of edits (that create exactly the new model from the old one). This problem can be reduced to searching for graph isomorphisms, a computationally hard task. Moreover, similarity degree does not always guarantee that one element in the new version can be tracked correctly back to its old version (there might exist ambiguities, at least).

The purpose of this thesis is to evaluate the current state-of-the-art model differencing approaches by measuring the performances of existing differencing approaches, especially when dealing with ontological information carried by the involved models. The work is expected to:
- investigate the commonly adopted strategies for carrying out differences between models;
- investigate the approaches available for representing ontological information in models;
- propose a benchmark approach to measure the performances of state-of-the-art differencing techniques, especially dealing with ontological information;
- realise a prototypical implementation of the benchmark algorithm enabling the measurement of the differencing technique results both performance-wise and precision-wise.
Student: Lorenzo Addazi addazi.lorenzo@gmail.com
Prerequisites: Knowledge of model-based development principles, of Eclipse EMF, and Java programming.
IDT supervisor: Antonio Cicchetti
antonio.cicchetti@mdh.se, +46-21-151762
Examinator: Jan Carlson
Jan Carlson
jan.carlson@mdh.se, +46-21-151722
Misc: Knowledge of ontology representation approaches is a plus.

Rapport och bilagor

Size

Senaste uppdatering


  • Mälardalen University |
  • Box 883 |
  • 721 23 Västerås/Eskilstuna |
  • 021-101300, 016-153600 |
  • webmaster |
  • Latest update: 2017.08.23