Prototyping a recommendation system for Ukiyo-e using hybrid recommendation algorithm

This paper introduces a recommendation system for Ukiyo-e prints. Although there are some services available online which recommends Ukiyo-e prints to users, they have not supported a recommendation function, yet. To meet the complex preferences of the users, the hybrid algorithm can fit as a feasible solution to realize the Ukiyo-e recommendation. In this paper, our prototype recommendation system for Ukiyo-e using it is proposed. Our proposed system adapts item-based collaborative filtering, user-based collaborative filtering and popularity-based filtering to predict which Ukiyo-e prints the user may like to view. Also this paper addresses the cold start problem and ways to profile users when limited authentication information is available.

[1]  Kyoji Kawagoe,et al.  Ukiyo-e Recommendation based on Deep Learning For Learning Japanese Art and Culture , 2017, ICISDM '17.

[2]  Shan Liu,et al.  Research of personalized news recommendation system based on hybrid collaborative filtering algorithm , 2016, 2016 2nd IEEE International Conference on Computer and Communications (ICCC).

[3]  Adam Prügel-Bennett,et al.  A Scalable, Accurate Hybrid Recommender System , 2010, 2010 Third International Conference on Knowledge Discovery and Data Mining.

[4]  Nazaraf Shah,et al.  Hybrid Recommendation System for Tourism , 2013, 2013 IEEE 10th International Conference on e-Business Engineering.

[5]  Jia Yu,et al.  The design of an open hybrid recommendation system for mobile commerce , 2008, 2008 11th IEEE International Conference on Communication Technology.

[6]  Paul Covington,et al.  Deep Neural Networks for YouTube Recommendations , 2016, RecSys.