Research Areas
SOFTWARE RE-ENGINEERING .
BUG-PREDICTION .
OBJECT-ORIENTED PROGRAM MODULARIZATION .
STATIC ANALYSIS OF SOFTWARE SYSTEMS .
SOFTWARE METRICS.
Research Description:
In today’s technologically motivated business world, evolution and maintenance of software systems are key processes carried out. over the decades. Long lasting software systems are inevitable for automating industrial devices as their longevity insures smoother and uninterrupted business operations. In theoretical terminology, software sustainability covers broad spectrum of measures needed to operate software system for longer time, i.e, stability of its infrastructure, adaptability to functional and environmental changes. However, there is still an opportunity to explore application of software metrics in broad spectrum of software quality and software sustainability, particularly during evolutionary phases of software development. As a matter of fundamental perspective of modularization, decay of architectural strength is often expected as the software longevity continues. Therefore, evaluation of architectural metrics during software evolution can provide comprehensive assessment of major sustainability concerns and quality oriented features. In particular, my research directions seek following main objectives:
Adding the evidence that source code based modularization metrics describing cohesion, coupling and programming practices can be effective enough to evaluate software systems quantitatively for better assessment of software quality.
Evidence that technical aspects of sustainability and software quality prescribed by modularization metrics ultimately help in understanding the composition of software systems
Software metrics can potentially help in prediction of quality features like, faults, bugs, source code violation, change prone components.
Proposed Framework:
This is framework that actually provides pictorial view of reliability prediction model.
Re-usability oriented package based source code design
Multi-pronged strategy to evaluate the fault-prediction models