Ticket triaging refers to the assignment of tickets in issue trackers to the most appropriate developer(s) for their solution. It is generally a time-consuming task in projects in which hundreds of tickets are created daily. The current thesis will deal with automatic bug triaging, that is the usage of machine learning techniques to analyze historical information about tickets and suggest the most appropriate developers for the solution. The proposed approach will be applied within a Czech-based company as an aid for the triaging process.

The first part of the thesis work will be based on the review of related works in the area of automated bug triaging, providing an overview of approaches that were tested successfully in past research.

The second part of the thesis will be dealing with the evaluation of several models and approaches of analysis, experimenting both with open source datasets and the company's dataset, providing a comparison of the results by means of the application of several models.

The final expected results from the thesis are i) a prototype based on a scripting infrastructure to train and test the models, ii) the experimental results with the performance indicators from several models and iii) a web front-end that could be used within the company to automate the process of bug triaging.