Prasenjit Banerjee

Work place: National Institute of Technology, Durgapur, West Bengal-713209, India

E-mail: prasenjitmca7@gmail.com

Website:

Research Interests: Computational Science and Engineering, Computational Engineering, Software Engineering, Data Structures and Algorithms, Formal Methods, Formal Semantics, Automata Theory, Formal Languages

Biography

Mr. Prasenjit Banerjee is presently a Ph.D. student of National Institute of Technology, Durgapur, under the supervision of Dr. Anirban Sarkar. His domain of research interest is Component Based Software Engineering and Formal Language and Automata Theory. He has about 9yrs teaching, research and industry experience. His total numbers of publications in various international platforms are about 20.

Author Articles
Quality Evaluation of Component-Based Software: An Empirical Approach

By Prasenjit Banerjee Anirban Sarkar

DOI: https://doi.org/10.5815/ijisa.2018.12.08, Pub. Date: 8 Dec. 2018

In recent days, component-based software engineering has become popular in the software industry for its reuse property. A suitable component-based software model is crucial for the effective design of the component-based software engineering. Quality assessment, evaluation, and analysis of a component model are highly essential to maintain the efficient design in the development of such system. Quality measurement for the component model will be more accurate, if it can be measured by a set of valid and meaningful metrics. This paper has proposed an empirical approach to validate a set of quality metrics along with a set of quality attributes for the design model of component-based software. In the proposed approach, metrics interdependencies have described using a Chi-Square non-parametric test. This paper has considered six different case studies of a well-known library management system to establish the metrics interdependency along with several quality attributes of a component model. This helps to identify the practically useful set of metrics for the quality assessment of high cohesive and low coupling metrics of the component-based system. A massive dataset has been collected from the 34 students of the institute on these six case studies. The Pearson's correlation method has been applied on the collected data set to identify the several correlations between the set of metrics and the set of quality attributes in terms of operation time. This facilitates to assess different crucial quality attributes of component-based system (CBS) design like complexity, analyzability, expressiveness etc.

[...] Read more.
Other Articles