A METHOD OF COMBINING DATA REDUCTION TECHNIQUES FOR EFFECTIVE BUG TRIAGE

In current scenario, software development is having challenge of handling bugs and optimizing software. Daily vast amount of bug information is generated and dealing with these software bugs is an important part in software industry to keep software upgraded. Programming associations spend enormous measure of expense on managing programming bugs as it is an inevitable step of settling bugs, which aims to adequately allot a developer to a new bug. Manual work constitutes huge time cost so to lessen this, text classification techniques are used to perform automatic bug triage.  For popular programming frameworks, the number of day by day submitted bug reports is high. Triaging these arriving reports is not only time consuming but also monotonous assignment. Bug triage by using software data reduction techniques means reduction of bug data set by keeping the originality and chooses appropriate developer for a bug to fix it. Hence for these purpose two software data lessening techniques Instance selection and Feature selection are used. Basically Instance selection technique is used to eliminate bug reports which contain similar information and Feature selection technique is used to remove non-informative words from the bug data set