Using Deep Learning for Positive Reviews Prediction in Explainable Recommendation Systems

In the recent years, recommender systems have begun to attract the attention of many online-based companies. While these systems are being developed to provide users with better recommendations, they suffer from the lack of explainability. The explainable recommendation systems are developed to solve the problem of why certain products or services are recommended to a particular user. However, less attention has been attracted for predicting positive reviews from the whole data in the context of explainable recommendation. Therefore, in this paper, we focus on developing a model that uses deep learning for predicting positive reviews in explainable recommendation systems. It enables users to get not only intuitive explanations for the recommended items, but also to get more transparency by investigating whether the explanations are positive ones. To evaluate the proposed model, we conduct experiments on a bench-mark dataset from Amazon. Experimental results demonstrate the efficacy of the proposed model against the baselines.