Abstract: Software Entropy Metrics for bug prediction have been validated on various software systems by different researchers. In our previous research, we have validated that Software Entropy Metrics calculated for Mozilla subsystem’s predict the future bugs reasonably well. In this study, the Software Entropy metrics are calculated for a subsystem of Android and it is noticed that these metrics are not suitable for bug prediction. The results are compared with a subsystem of Mozilla and a comparison is made between the two software systems to determine the reasons why Software Entropy metrics are not applicable for Android.
Abstract: The aim of this paper is to rank the impact of Object
Oriented(OO) metrics in fault prediction modeling using Artificial
Neural Networks(ANNs). Past studies on empirical validation of
object oriented metrics as fault predictors using ANNs have focused
on the predictive quality of neural networks versus standard
statistical techniques. In this empirical study we turn our attention to
the capability of ANNs in ranking the impact of these explanatory
metrics on fault proneness. In ANNs data analysis approach, there is
no clear method of ranking the impact of individual metrics. Five
ANN based techniques are studied which rank object oriented
metrics in predicting fault proneness of classes. These techniques are
i) overall connection weights method ii) Garson-s method iii) The
partial derivatives methods iv) The Input Perturb method v) the
classical stepwise methods. We develop and evaluate different
prediction models based on the ranking of the metrics by the
individual techniques. The models based on overall connection
weights and partial derivatives methods have been found to be most
accurate.
Abstract: Missing data is a persistent problem in almost all
areas of empirical research. The missing data must be treated very
carefully, as data plays a fundamental role in every analysis.
Improper treatment can distort the analysis or generate biased results.
In this paper, we compare and contrast various imputation techniques
on missing data sets and make an empirical evaluation of these
methods so as to construct quality software models. Our empirical
study is based on NASA-s two public dataset. KC4 and KC1. The
actual data sets of 125 cases and 2107 cases respectively, without
any missing values were considered. The data set is used to create
Missing at Random (MAR) data Listwise Deletion(LD), Mean
Substitution(MS), Interpolation, Regression with an error term and
Expectation-Maximization (EM) approaches were used to compare
the effects of the various techniques.
Abstract: Importance of software quality is increasing leading to development of new sophisticated techniques, which can be used in constructing models for predicting quality attributes. One such technique is Artificial Neural Network (ANN). This paper examined the application of ANN for software quality prediction using Object- Oriented (OO) metrics. Quality estimation includes estimating maintainability of software. The dependent variable in our study was maintenance effort. The independent variables were principal components of eight OO metrics. The results showed that the Mean Absolute Relative Error (MARE) was 0.265 of ANN model. Thus we found that ANN method was useful in constructing software quality model.