Bug-fixing, or fixing known errors in computer software, makes up a large portion of software development expenses. Once a bug is discovered, it must be assigned to an appropriate developer who has the necessary expertise to fix the bug. This bug-assignment task has traditionally been done manually. However, this manual task is time-consuming, error-prone, and tedious. Therefore, automatic bug assignment techniques have been developed to facilitate this task. Most of the existing techniques are report-based. That is, they work on bugs that are textually described in bug reports. However, only a subset of bugs that are observed as a faulty program execution are also described textually. Certain bugs, such as security vulnerability bugs, are only represented with a faulty program execution, and are not described textually. In other words, these bugs are represented by a code coverage, which indicates which lines of source code have been executed in the faulty program execution. Promptly fixing these software security vulnerability bugs is necessary in order to manage security threats. Accordingly, execution-based bug assignment techniques, which model a bug with a faulty program execution, are an important tool in fixing software security bugs. In this thesis, we compare WhoseFault, an existing execution-based bug assignment technique, to report-based techniques. Additionally, we propose DR_BEV (Developer Recommendation Based on Executed Vocabulary), a novel execution-based technique that models developer expertise based on the vocabulary of each developer's source code contributions, and we demonstrate that this technique out-performs the current state-of-the-art execution-based technique. Our observations indicate that report-based techniques perform better than execution-based techniques, but not by a wide margin. Therefore, while a report-based technique should be used if a report exists for a bug, our results should provide confidence in the scenarios in which only execution-based techniques are applicable. / Master of Science / Bug-fixing, or fixing known errors in computer software, makes up a large portion of software development expenses. Once a bug is discovered, it must be assigned to an appropriate developer who has the necessary expertise to fix the bug. This bug-assignment task has traditionally been done manually. However, this manual task is time-consuming, error-prone, and tedious. Therefore, automatic bug assignment techniques have been developed to facilitate this task. Most of the existing techniques are report-based. That is, they work on bugs that are textually described in bug reports. However, only a subset of bugs that are observed as a faulty program execution are also described textually. Certain bugs, such as security vulnerability bugs, are only represented with a faulty program execution, and are not described textually. In other words, these bugs are represented by a code coverage, which indicates which lines of source code have been executed in the faulty program execution. Promptly fixing these software security vulnerability bugs is necessary in order to manage security threats. Accordingly, execution-based bug assignment techniques, which model a bug with a faulty program execution, are an important tool in fixing software security bugs. In this thesis, we compare WhoseFault, an existing execution-based bug assignment technique, to report-based techniques. Additionally, we propose DR_BEV (Developer Recommendation Based on Executed Vocabulary), a novel execution-based technique that models developer expertise based on the vocabulary of each developer's source code contributions, and we demonstrate that this technique out-performs the current state-of-the-art execution-based technique.

In software development projects, bug triage consists mainly of assigning bug reports to software developers or teams (depending on the project). The partial or total automation of this task would have a positive economic impact on many software projects. This thesis introduces a systematic four-step method to find some of the best configurations of several machine learning algorithms intending to solve the automatic bug assignment problem. These four steps are respectively used to select a combination of pre-processing techniques, a bug report representation, a potential feature selection technique and to tune several classifiers. The aforementioned method has been applied on three software projects: 66 066 bug reports of a proprietary project, 24 450 bug reports of Eclipse JDT and 30 358 bug reports of Mozilla Firefox. 619 configurations have been applied and compared on each of these three projects. In production, using the approach introduced in this work on the bug reports of the proprietary project would have increased the accuracy by up to 16.64 percentage points.