Automatically identifying and understanding students' questions aboutproblems they encounter or issues related to their universities is very important foruniversities to promptly grasp the aspirations of their students. This enables them tosupport and satisfy their students and enhance their reputation. Especially as socialnetworks and online media continue to develop, students can easily post theirquestions and concerns online. This makes it easier for universities to access andaddress student questions. Although this is not a new problem, it still faces manychallenges due to issues in natural language processing. To address this problem,within the scope of this article, we conduct a survey, perform experiments, andpropose a model to automatically classify students' questions into 11 areas of interestat the University of Transport and Communications. We conducted carefulexperiments with a dataset of more than ten thousand posts collected from websites,forums, and school fan pages. Finally, we obtained a model with prediction resultsthat achieved an accuracy of over 85%.Automatically identifying and understanding students' questions aboutproblems they encounter or issues related to their universities is very important foruniversities to promptly grasp the aspirations of their students. This enables them tosupport and satisfy their students and enhance their reputation. Especially as socialnetworks and online media continue to develop, students can easily post theirquestions and concerns online. This makes it easier for universities to access andaddress student questions. Although this is not a new problem, it still faces manychallenges due to issues in natural language processing. To address this problem,within the scope of this article, we conduct a survey, perform experiments, andpropose a model to automatically classify students' questions into 11 areas of interestat the University of Transport and Communications. We conducted carefulexperiments with a dataset of more than ten thousand posts collected from websites,forums, and school fan pages. Finally, we obtained a model with prediction resultsthat achieved an accuracy of over 85%.