The recent developments in the software technology assist humanities in various fields including engineering, technology, management, medical science, research, education, banking etc. Fault identification is a crucial one to the software testing professionals since a huge number of tests are carried out to identify the level of the defect. Therefore the machine learning algorithms are employed to develop software fault detection model in order to predict the fault in the software. The irrelevant and redundant test data reduces the accuracy of fault detection model. The accuracy of the fault detection model highly depends on the number of significant relevant test data. Therefore feature selection concept is applied to select the accurate features for developing the fault detection model. This paper proposes a method to select appropriate features with honey bee optimization technique for reducing the search space and to improve the accuracy in the software fault detection.
General TermsIntroduction to the feature selection, applications of feature selection, software fault detection using feature selection, data preprocessing for improving accuracy in supervised machine learning algorithm, application of data mining, honey bee optimization in data mining approaches.
KeywordsFeature selection, software fault detection algorithm, honey bee optimization, data mining approaches, machine learning algorithm, improving accuracy in classification algorithms.