POSTPONED - "The structure of social influence in recommender networks" -- talk by Pantelis P. Analytis (University of Southern Denmark) Event

Event details

The ability of people to influence the opinion of others on matters of taste varies greatly---both in the offline world and in recommender systems. What are the mechanisms underlying this striking inequality? We use the weighted k-nearest-neighbor algorithm to represent an array of social learning strategies and show---using network theory---how this gives rise to networks of social influence in six real-world domains of taste. By doing so, we show three novel results that apply both to offline advice taking and online recommender settings. First, influential individuals have mainstream tastes and high dispersion in their taste similarity with others. Second, the fewer people an individual or algorithm consults (i.e., the lower k) and the more sensitive an individual or algorithm is to how similar other people are, the smaller the group of people with substantial influence. Third, the influence networks that emerge are hierarchically organized. Our results shed new light on classic empirical findings in communication and network science and can help improve our understanding of social influence in the offline and online world.