Collaborative Filtering via Implicit Culture

Implicit Culture is the relation existing between a set and a group of agents such that the elements of the set behave according to the culture of the group. Earlier work claimed that supporting Implicit Culture phenomena can be useful for both artificial and human agents. In particular, collaborative filtering - the popular technique exploited in reccomendation systems - can be seen as a System for Implicit Culture Support (SICS). In this paper, we refine the definition of Implicit Culture, present an implementation of a SICS, and show its functional equivalence to collaborative filtering on the particular domain of item-ranking agent with an adequate a priori theory. Results and advantages of the more general framework are presented and discussed