Internet recommendation systems

Abstract:

A study examines the merits of methods used by online firms such as Yahoo!, Amazon.com and Movie Critic in recommending documents and products to consumers. It is suggested that preference models used in marketing offer good alternatives, and describe a Bayesian preference model that allows statistical integration of five types of information useful for making recommendations: 1. a person's expressed preferences, 2. preferences of other consumers, 3. expert evaluations, 4. item characteristics, and 5. individual characteristics. The proposed method accounts for not only preference heterogeneity across users but also unobserved product heterogeneity by introducing the interaction of unobserved product attributes with customer characteristics. Copyright American Marketing Association Aug 2000