Optimizing the Design of Recommendation Systems
Xiaohan Guo, Dejun Xie · 2021
With the development of digital technology and internet applications, the massive market data and information overload make it hard for users to organize and make good use of comments to choose their preferred items. In this connection, recommendation systems are widely developed with high expectations. While there have been substantial researches focusing on the recommendation methods and implementations, this paper provides an up-to-date synthesis of the essential factors towards a more rigorous, user-helping system design in consolidation of collaborative filtering, content-based model, and hybrid model in a fast-developing web and social network environment. Aspects of evaluation metrics are analyzed and compared in connection to particular applicabilities in the system design. Further explorations in terms of credit rating are discussed in a real-business setting.