Preface to the special issue on context-aware recommender systems

A comprehensive survey and analysis of the state of the art on time-aware recommender systems (TARS) is discussed. An empirical comparison was conducted of the impact of several evaluation protocols on measuring relative performances of three widely used TARS approaches and one well-known non-contextual recommendation approach. The results obtained showed that the use of different evaluation conditions not only yields remarkable differences between metrics measuring distinct recommendation properties. They also may affect the relative ranking of approaches for a particular metric. The recommendation problem relies on the notion of ratings as a mechanism to capture user preferences for different items. The rating values express a scale of the users' preferences for the items, or are related to the users' consumption of items. The recommendation problem can be reduced to solve a rating prediction problem, which consists of predicting unknown ratings for pairs by providing an approximation of the function.