As Software’s are fault prone and there is a need to predetermine the chances of existence of bugs while developing the software on the basis of Different Metrics/Factors which are important. Through Bug prediction models, Software developers can know in advance which factors are important so that they will make sure that Probability of coming of Software Bugs post Release/Implementation is minimal/least. This Research paper emphasizes mainly on prediction of the software post release bugs for the Eclipse Software. In order to reduce the number of dimensions of the input feature vector, factor analysis which uses concept of feature selection is applied and new matrix with lower number of dimensions is used as input to general linear regression based prediction models. Finally results are compared among different versions of Eclipse i.e., version 2.0 & 2.1 with correlation based dimension selection process and empirical study was conducted to compare prediction results.
Impact Factor
2022: 6.012
This work is licensed under a