Improving automatic bug assignment using time‐metadata in term‐weighting

Abstract

Assigning newly reported bugs to project developers is a time-consuming and tedious task for triagers using the traditional manual bug triage process. Previous efforts for creating automatic bug assignment systems use machine learning and information-retrieval techniques. These approaches commonly use tf-idf, a statistical computation technique for weighting terms based on term frequency. However, tf-idf does not consider the metadata, such as the time frame at which a term was used, when calculating the weight of the terms. This study proposes an alternate term-weighting technique to improve the accuracy of automatic bug assignment approaches that use a term-weighting technique. This technique includes the use of metadata in addition to the statistical computation to calculate the term weights. Moreover, it restricts the set of terms used to only nouns. It was found that when using only nouns and the proposed term-weighting technique, the accuracy of an automatic bug assignment approach improves from 12 to 49% over tf-idf for three open-source projects.

Investigating the information value of different sources of evidence of developers’ expertise for bug assignment in open‐source projects

“…In a recent research [17], we proposed a method called VTBA which is based on original TF‐IDF [18, 27–29] and showed that it outperforms many other recent approaches. It only considers the technical keywords (the ones that are a Stack Overflow tag) and disregards others.…”

“…We report its results after the final run on all projects (excluding the three test projects), with the configuration we obtained from our test projects. In order to validate our approach, we implemented three previous approaches; the baseline TF‐IDF [26, 27, 29], Time‐TF‐IDF [18] and our previous approach, VTBA [17]. We tried our best to implement those approaches as precise as possible.…”

Survey Based Classification of Bug Triage Approaches

“…To reduce the time and cost of bug assignment process, the first automatic bug triager was proposed by Cubranic and Murphy [7]. Thereafter, many automatic bug triage approaches were proposed that are based on machine learning [8][9][10][11][12][13][14][15][16], metadata [17][18][19][20][21][22][23], or developer profile [24][25][26][27]. These are shown in Table 1.…”

“…Assigning change requests to software developers [8] Automatic assignment of work item [9] Reducing the Effort of Bug Report Triage: Recommenders for Development-Oriented Decisions [10] Highly-accurate Bug Triage using Machine Learning [11] Improving bug triage with Bug tossing Graphs [12] Automated, highly-accurate, bug assignment using machine learning and tossing graphs [13] An Approach to Improving Bug Assignment with bug tossing graph and bug similarities [14] Novel metrics for bug triage [15] Automatic Bug Triage using Semi-Supervised Text Classification [16] Meta-Data Based Approch COSTRIAGE: A Cost-Aware Triage Algorithm for Bug Reporting [17] A time based approach to Automatic Bug Report Assignment [18] Topic-based, time aware bug assignment [19] Improving automatic bug assignment using time-meta in term weights [20] Effective Bug Triage based on Historical Bug-Fix information [21] Automatic Bug Assignment Using Information Extraction Methods [22] A Noun based approach to feature location using time aware term-weighting […”

“…In some of the projects, few of the developers" works as a gate keepers who have only the permission to commit to the source code, that means developer who changes the source code and who committed the change(s) in the software repository, can be different. [20] developer" recommendation system based on the similarity (weight) between the new bug" term and previous bugreport information (corpus). They considered important details from a bug like: the term creation, modification time and who changed it.…”