Hybrid recommender system with temporal information
Farman Ullah, Ghulam Sarwar, Sung Chang Lee, Yun Kyung Park, Kyeong Deok Moon, Jin Tae Kim · 2012
In the last few years, many recommender systems have been proposed but most of them suffer from scalability, sparsity and cold start issues. The existing recommender systems don't consider contextual information in term of user current device, location, company and time etc. In this paper, we proposed Hybrid Recommender System that accounts item attributes similarity, user rating similarity, user demographic similarity and the temporal information to do recommendation. The proposed algorithm will produce better results as it uses temporal information in computing and uses hybrid structure, model-based and memory-based system to improve system scalability and accuracy simultaneously. It uses the temporal information in the recommendation process to make recommendation for user at specific time.