Software Quality Prediction Using Fuzzy Logic Technique

Software Quality Prediction Using Fuzzy Logic Technique

Saumendra Pattnaik (Siksha'O'Anusandhan University, Odisha, India), Binod Kumar Pattanayak (Siksha'O'Anusandhan University, Odisha, India) and Srikanta Patnaik (Siksha'O'Anusandhan University, Odisha, India)
Copyright: © 2019 |Pages: 21
DOI: 10.4018/IJISSS.2019040104

Abstract

In the global market, the value of software quality is a major factor in software industries. In a majority of cases, software quality needs to be quantified and measured. Previously, many models like the ISO model, the McCall model, the Boehm model, etc., have been used for the quantification of the parameters which determine the software quality. The software quality can be judged by analyzing all the characteristics of the product. In this article, a new pragmatic model which is the modification of the existing ISO Model has been proposed in order for the quantification of the software quality factors in a more precise way. Because of the nature of the attributes which determine the software quality, a Fuzzy logic-based approach is considered to be a better technique among various existing machine learning techniques. Here, the simulink tool is used for constructing the running model which helps the net quality of the software.
Article Preview
Top

In the year 2012, authors Pasrija et al. (Pasrija & Kumar, 2012) in their research paper “Assessment of Software Quality: Choquet Integral Approach,” proposed a unique methodology that can be successfully used for comparison of different solutions related to software quality assessment on the basis of the SRS document of a common problem. The Choquet integral Approach is used here for comparing different salutations for software using Fuzzy measure identification and decision making through rate of criteria. In this approach, it aggregates several values to a single value. In previous approaches, the software quality was quantified but it did not take into consideration the Fuzzy nature of the parameters pertaining to software quality assessment.

Some more approaches have been worked out relating to software quality analysis using “Ensemble of software defect predictors: an AHP-based evaluation method”, by Peng, Yi et al. (Peng & Yi, 2011) in the year 2011. The same work using weighted arithmetic mean was extended by Srivastava et al. (Srivastava, Jain, Singh & Raghurama, 2009) in the year of 2009.

An approach for evaluating the prepared alternative solution is done by Vatesh et al. (Grabisch & Roubens, 2000) in the year 2012 based on software requirement specification (SRS) of a given problem. In this approach, the authors use ISO model for selecting the criteria those can be usefully taken into consideration for assessment of software quality. The Fuzzy measure shows the importance of a particular criteria or a group of criteria when combined together. An importance of a criterion is calculated from its membership values. In this paper, the approach of Choquet integral is used by aggregating the values of manufacturing view, user view and product view ignoring its final value.

As per authors Yuen et al. (Yuen & Lau, 2008) in the year 2008, a discussion is made about Analytic Hierarchical Process (AHP) which is an aggregation technique. For avoiding the limitation of this approach, they have introduced the Fuzzy Analytic Hierarchical Process (Ravasan & Mansouri, 2015). Here authors propose a model for prediction of software quality that is based on Falcon model. It is based on the technology like Artificial Neural Network, Computational Intelligence and Fuzzy logic. In this approach, the model is capable of dealing with the objective data collected during the software development process. For avoiding the limitation of AHP they have extended it to Fuzzy AHP. In this paper, they used triangular fuzzy number for better software quality prediction.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 12: 4 Issues (2020): 1 Released, 3 Forthcoming
Volume 11: 4 Issues (2019)
Volume 10: 4 Issues (2018)
Volume 9: 4 Issues (2017)
Volume 8: 4 Issues (2016)
Volume 7: 4 Issues (2015)
Volume 6: 4 Issues (2014)
Volume 5: 4 Issues (2013)
Volume 4: 4 Issues (2012)
Volume 3: 4 Issues (2011)
Volume 2: 4 Issues (2010)
Volume 1: 4 Issues (2009)
View Complete Journal Contents Listing