In this paper we introduce a mathematical model that captures some of the salient features of recommender systems that are based on popularity and that try to exploit social ties among the users. We show that, under very general conditions, the market always converges to a steady state, for which we are able to give an explicit form. Thanks to this we can tell rather precisely how much a market is altered by a recommendation system, and determine the power of users to influence others. Our theoretical results are complemented by experiments with real world social networks showing that social graphs prevent large market distortions in spite of the presence of highly influential users.
Description
The Limits of Popularity-Based Recommendations, and the Role of Social Ties
%0 Conference Paper
%1 Bressan:2016:LPR:2939672.2939797
%A Bressan, Marco
%A Leucci, Stefano
%A Panconesi, Alessandro
%A Raghavan, Prabhakar
%A Terolli, Erisa
%B Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining
%C New York, NY, USA
%D 2016
%I ACM
%K popularity-based-recommendation social-information-access social-ties
%P 745--754
%R 10.1145/2939672.2939797
%T The Limits of Popularity-Based Recommendations, and the Role of Social Ties
%U http://doi.acm.org/10.1145/2939672.2939797
%X In this paper we introduce a mathematical model that captures some of the salient features of recommender systems that are based on popularity and that try to exploit social ties among the users. We show that, under very general conditions, the market always converges to a steady state, for which we are able to give an explicit form. Thanks to this we can tell rather precisely how much a market is altered by a recommendation system, and determine the power of users to influence others. Our theoretical results are complemented by experiments with real world social networks showing that social graphs prevent large market distortions in spite of the presence of highly influential users.
%@ 978-1-4503-4232-2
@inproceedings{Bressan:2016:LPR:2939672.2939797,
abstract = {In this paper we introduce a mathematical model that captures some of the salient features of recommender systems that are based on popularity and that try to exploit social ties among the users. We show that, under very general conditions, the market always converges to a steady state, for which we are able to give an explicit form. Thanks to this we can tell rather precisely how much a market is altered by a recommendation system, and determine the power of users to influence others. Our theoretical results are complemented by experiments with real world social networks showing that social graphs prevent large market distortions in spite of the presence of highly influential users.},
acmid = {2939797},
added-at = {2017-02-08T21:10:39.000+0100},
address = {New York, NY, USA},
author = {Bressan, Marco and Leucci, Stefano and Panconesi, Alessandro and Raghavan, Prabhakar and Terolli, Erisa},
biburl = {https://www.bibsonomy.org/bibtex/2dcf986d4a58ece7d34e848fe66d2d838/xianteng},
booktitle = {Proceedings of the 22Nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining},
description = {The Limits of Popularity-Based Recommendations, and the Role of Social Ties},
doi = {10.1145/2939672.2939797},
interhash = {f221712bd6c530346f78fd98cd52776a},
intrahash = {dcf986d4a58ece7d34e848fe66d2d838},
isbn = {978-1-4503-4232-2},
keywords = {popularity-based-recommendation social-information-access social-ties},
location = {San Francisco, California, USA},
numpages = {10},
pages = {745--754},
publisher = {ACM},
series = {KDD '16},
timestamp = {2017-02-08T21:10:39.000+0100},
title = {The Limits of Popularity-Based Recommendations, and the Role of Social Ties},
url = {http://doi.acm.org/10.1145/2939672.2939797},
year = 2016
}