Effective bug triage with Prim's algorithm for feature selection

Abstract

For constructing any software application or item it is essential to identify the bug in the item while building up the product. At every stage of testing the bug report is created, more of the time is wasted for settling the bug. In settling the bug, software enterprises waste 45 percent of cost. One of the basic systems for settling the bug is bug triage. It is a process for settling the bugs whose fundamental object is to properly allocate a designer to a novel bug for further taking handling. If the assigned developer is busy or not available then this system assigns a new domain specific developer. Initially manual work is accomplished for each time creating the bug report. After that content analysis strategies are functional to conduct normal bug triage. The current framework resists the problem of data reduction in the settling of bugs naturally. Furthermore there is a need of methods which decreases the range likewise enhances the excellence of bug data. For feature collection which is not given precise outcome traditional framework utilized CHI technique. In this way this framework proposed the technique for feature selection by utilizing the Prim's strategy. By joining the instance selection and the feature selection calculations to simultaneously diminish the data scale likewise upgrade precision of the bug reports in the bug triage. By utilizing Prim's strategy, noisy words are removed from the dataset set. From the experimental result it is displayed that accuracy of the proposed system is greater than the accuracy of the existing system.