Research output

Article in Journal ()

Improved Binary Similarity Measures for Software Modularization

Citation
Naseem R, Deris MBM, Maqbool O, Li J, Shahzad S & Shah H (2017) Improved Binary Similarity Measures for Software Modularization, Frontiers of Information Technology and Electronic Engineering, 18 (8), pp. 1082-1107.

Abstract
Various binary similarity measures have been employed in clustering approaches to make homogeneous groups of similar entities in the data. These similarity measures are mostly based only on the presence and absence of features. Binary similarity measures have also been explored with different clustering approaches (e.g., agglomerative hierarchical clustering) for software modularization to make the software systems understandable and manageable. Each similarity measure has its own strengths and weaknesses that result in improving and deteriorating the clustering results, respectively. This paper highlights the strengths of some well-known existing binary similarity measures for software modularization. Furthermore, based on these existing similarity measures, this paper introduces the improved new binary similarity measures. Proofs of the correctness with illustration and a series of experiments are presented to evaluate the effectiveness of our new binary similarity measures.

Keywords
Binary similarity measure; Binary features; Combination of measures; Software modularization

StatusPublished
AuthorsNaseem Rashid, Deris Mustafa Bin Mat, Maqbool Onaiza, Li Jingpeng, Shahzad Sarah, Shah Habib
Publication date08/2017
Publication date online22/09/2017
Date accepted by journal12/04/2016
PublisherSpringer
ISSN 2095-9184
LanguageEnglish

Journal
Frontiers of Information Technology and Electronic Engineering: Volume 18, Issue 8

© University of Stirling FK9 4LA Scotland UK • Telephone +44 1786 473171 • Scottish Charity No SC011159
My Portal