I am a PhD student at USC 's Information Sciences Institute . I work with prof. Aram Galstyan and prof. Nanyun Peng in Machine INtelligence and Data Science group (MINDS) and Peng's Language Understanding & Synthesis Lab (PLUS) labs. I am interested in Natural Language Processing specifically Dialogue Systems and their evaluations.
Recent advances in open domain dialogue systems highlight the difficulties in automatically evaluating them. In this project, we try to come up with automatic evaluation metrics which evaluate the quality of responses generated by dialogue systems from different aspects such as relevancy and engagement that can be good substitutions for human evalautions.
Memory-based collaborating filtering techniques are widely used in recommender systems. They are based on full initial ratings in a user-item matrix. However, most of the time in group recommender systems, this matrix is sparse. This deficiency may make memory-based collaborative filtering unsuitable for group recommender systems. In this paper, we tried to improve memory-based techniques for group recommendation systems by resolving the data sparsity problem.