Aspect category detection is one challenging subtask of aspect based sentiment analysis, which categorizes a review sentence into a set of predefined aspect categories. Most existing methods regard the aspect category detection as a flat classification problem. However, aspect categories are inter-related, and they are usually organized with a hierarchical tree structure. To leverage the structure information, this paper proposes a hierarchical multi-label classification model to detect aspect categories and uses a graph enhanced transformer network to integrate label dependency information into prediction features. Experiments have been conducted on four widely-used benchmark datasets, showing that the proposed model outperforms all strong baselines.

Community Question Answering (CQA) in web forums, as a classic forum for user communication, provides a large number of high-quality useful answers in comparison with traditional question answering. Development of methods to get good, honest answers according to user questions is a challenging task in natural language processing. Many answers are not associated with the actual problem or shift the subjects, and this usually occurs in relatively long answers. In this paper, we enhance answer selection in CQA using multi-dimensional feature combination and similarity order. We make full use of the information in answers to questions to determine the similarity between questions and answers, and use the text-based description of the answer to determine whether it is a reasonable one. Our work includes two subtasks: (a) classifying answers as good, bad, or potentially associated with a question, and (b) answering YES/NO based on a list of all answers to a question. The experimental results show that our approach is significantly more efficient than the baseline model, and its overall ranking is relatively high in comparison with that of other models.