Abstract:
This thesis proposes an automated approach to classifying questions that are posted on Stack Overflow website with regard to a certain kind of database products in particular. Such information is valuable to database product owners for improving their products. The categories of questions are defined at two levels, i.e. problem and subproblem. The problem level includes development, installation, and performance tuning, while the subproblem level consists of design, limitation, and discussion. By cross-combining the two levels, questions can be classified into nine problem-subproblem classes. Natural language processing and text classification are used with several machine learning algorithms. The best classifier for all classes is used in a web application that can classify each question by a problem-subproblem tag. In addition, all classified questions are further analyzed by using a topic modeling algorithm to identify the topics that are addressed in those questions. This will be additional information for a database product owner to understand the issues of the database product for further improvement.