@inproceedings{Rosas:2018,
author = {Rosas, F and Chen, K-C and Gunduz, D},
title = {Social diversity for reducing the impact of information cascades on  social learning},
url = {http://arxiv.org/abs/1805.05500v1},
year = {2018}
}

Download

TY  - CPAPER
AB  - Collective behavior in online social media and networks is known to becapable of generating non-intuitive dynamics associated with crowd wisdom andherd behaviour. Even though these topics have been well-studied in socialscience, the explosive growth of Internet computing and e-commerce makes urgentto understand their effects within the digital society. In this work we explorehow the stochasticity introduced by social diversity can help agents involvedin a inference process to improve their collective performance. Our resultsshow how social diversity can reduce the undesirable effects of informationcascades, in which rational agents choose to ignore personal knowledge in orderto follow a predominant social behaviour. Situations where social diversity isnever desirable are also distinguished, and consequences of these findings forengineering and social scenarios are discussed.
AU  - Rosas,F
AU  - Chen,K-C
AU  - Gunduz,D
PY  - 2018///
TI  - Social diversity for reducing the impact of information cascades on  social learning
UR  - http://arxiv.org/abs/1805.05500v1
ER  - 

Download