For large scale software systems, many bugs can be reported over a long period of time. For software quality assurance and software project management, it is important to assign adequate resources to resolve the reported bug. An important issue concerning assignment is the ability to predict bug-fixing time because it can help a project team better estimate software maintenance efforts and better manage software projects. In this paper, we propose a model that can predict the bug-fixing time using the text information extracted from Bugzilla, an on-line open source Bug Tracking System (BTS). We perform an empirical investigation for the bugs of Novell, OpenOffice and LiveCode, three open source projects using Bugzilla. Proposed model is based on historical data stored on the BTS. For each bug-report we build a classification model to predict the time of its resolution, as slow or fast. In this work we used, as classifier, Support Vector Machine (SVM) but different classifier can be easily used. Our model, differently from existing work reported in the literature, selects all and only the attributes useful for prediction and filters appropriately attributes for the test-set. Experimental results show the model is effective. In the future, we will use and compare other different classification method to select the best one for a specific data-set.
Knowledge extraction from on-line open source bug tracking systems to predict bug-fixing time
ARDIMENTO, PASQUALE;
2017-01-01
Abstract
For large scale software systems, many bugs can be reported over a long period of time. For software quality assurance and software project management, it is important to assign adequate resources to resolve the reported bug. An important issue concerning assignment is the ability to predict bug-fixing time because it can help a project team better estimate software maintenance efforts and better manage software projects. In this paper, we propose a model that can predict the bug-fixing time using the text information extracted from Bugzilla, an on-line open source Bug Tracking System (BTS). We perform an empirical investigation for the bugs of Novell, OpenOffice and LiveCode, three open source projects using Bugzilla. Proposed model is based on historical data stored on the BTS. For each bug-report we build a classification model to predict the time of its resolution, as slow or fast. In this work we used, as classifier, Support Vector Machine (SVM) but different classifier can be easily used. Our model, differently from existing work reported in the literature, selects all and only the attributes useful for prediction and filters appropriately attributes for the test-set. Experimental results show the model is effective. In the future, we will use and compare other different classification method to select the best one for a specific data-set.I documenti in IRIS sono protetti da copyright e tutti i diritti sono riservati, salvo diversa indicazione.