A Survey of Recommender System from Data Sources Perspective
Huaiyu Pi, Zhenyan Ji, Yang Chun · Proceedings of the 2018 8th International Conference on Management, Education and Information (MEICI 2018) · 2018
In order to solve the problem of information overload in big data era, the personalized recommender system has been widely used.Collaborative filtering, as a classical algorithm, has become the basis of the recommender system.In recent years, there are more and more recommender systems based on multiple data sources are proposed.Today's recommender systems integrate multiple data sources and recommendation methods are more accurate and explainable compare with rating-based recommendation systems.How to integrate multiple data sources to further improve the accuracy and interpretability of recommendation results, reduce computational complexity and cold start risk has become the key content of recommendation researches.