Experimenting Emotion-Based Book Recommender Systems with Social Data

In this contribution, we present two methods for book recommendations incorporating emotions extracted from online reviews, using two distinct recommender systems techniques: Content-based filtering and Collaborative filtering. This paper is based on a previous conference publication [14] and extends the system description and experimental setup. The recommender systems are experimentally validated using three datasets of different sizes, collected from Goodreads website - a popular book social network, using our customized web scraper. Lastly, we propose and use two evaluation metrics: Coverage and Average Recommendations Similarity, and discuss our results.