Balaji Satya Pradeep Veera: Multi-Objective Issue Tracking Framework. Otto-von-Guericke-Unviersität Magdeburg, 2017.

Abstract

Assigning an issue reports to individual developers is typically a manual, time-consuming and tedious task for the triager. Many automated issue tracking systems have been proposed over the time to choose an optimal developer for an issue report. These Single-Objective approaches try to find a developer who has fixed similar issue reports in the past. Thus they find an optimal developer for an issue report. While choosing an optimal developer, the Single-Objective approaches don’t consider the severity of
the issue report and the workload of the developer. The assigned issue report might reassign to another developer if the first assigned developer is busy with other issue reports. To solve this problem, we propose a framework of automated issue tracking system called Multi-Objective Issue Tracking Framework (MOITF). MOITF considers the workload of the developer along with the severity of the issue report while choosing the developer. This approach analyzes the history of previously closed issue reports for making predictions to the new issue report. The problem of selecting a developer who has fixed similar issue reports in the past and had less workload is viewed as a Multi objective problem. The traditional non-dominated sorting algorithm is used to find non-dominated devel-
opers who have knowledge about the problem mentioned in the issue report and also have time to fix the issue report.

    BibTeX (Download)

    @mastersthesis{Veera2017,
    title = {Multi-Objective Issue Tracking Framework},
    author = {Balaji Satya Pradeep Veera},
    editor = {Sebastian Nielebock and Frank Ortmeier},
    year  = {2017},
    date = {2017-07-28},
    school = {Otto-von-Guericke-Unviersit\"{a}t Magdeburg},
    abstract = {Assigning an issue reports to individual developers is typically a manual, time-consuming and tedious task for the triager. Many automated issue tracking systems have been proposed over the time to choose an optimal developer for an issue report. These Single-Objective approaches try to find a developer who has fixed similar issue reports in the past. Thus they find an optimal developer for an issue report. While choosing an optimal developer, the Single-Objective approaches don’t consider the severity of
    the issue report and the workload of the developer. The assigned issue report might reassign to another developer if the first assigned developer is busy with other issue reports. To solve this problem, we propose a framework of automated issue tracking system called Multi-Objective Issue Tracking Framework (MOITF). MOITF considers the workload of the developer along with the severity of the issue report while choosing the developer. This approach analyzes the history of previously closed issue reports for making predictions to the new issue report. The problem of selecting a developer who has fixed similar issue reports in the past and had less workload is viewed as a Multi objective problem. The traditional non-dominated sorting algorithm is used to find non-dominated devel-
    opers who have knowledge about the problem mentioned in the issue report and also have time to fix the issue report.},
    keywords = {Framework, Issue Tracking, multi-objective},
    pubstate = {published},
    tppubtype = {mastersthesis}
    }