Reviews: Towards Deep Conversational Recommendations

Summary: This paper provides a large-scale data set of 10,000 conversations for the domain of movie recommendation. Strengths: The authors will release the first large-scale dataset for conversational movie recommendations, which can be significant for follow-up work on conversational recommendation. The authors propose an end-to-end trainable architecture, with sub-components of new neural-based models for sentiment analysis and cold-start recommendations, along with natural language. Weaknesses: - As this work has the perspective of task-oriented recommendation, it seems that works such as [] Li, Xiujun, et al. "End-to-end task-completion neural dialogue systems." Also, discussion in general on how their work differs from other chatbox research works e.g.