Friends for Free: Self-Organizing Artificial Social Networks for Trust and Cooperation

TL;DR

This paper introduces SLACER, an algorithm enabling peers to self-organize artificial social networks that foster trust and cooperation, improving performance in decentralized systems without relying on natural social networks.

Contribution

The paper presents SLACER, a novel algorithm for creating and managing artificial social networks that are connected, cooperative, and robust, mimicking human friendship properties.

Findings

ASN are connected and cooperative

Networks exhibit trust between friends

Networks are robust and have short paths

Abstract

By harvesting friendship networks from e-mail contacts or instant message "buddy lists" Peer-to-Peer (P2P) applications can improve performance in low trust environments such as the Internet. However, natural social networks are not always suitable, reliable or available. We propose an algorithm (SLACER) that allows peer nodes to create and manage their own friendship networks. We evaluate performance using a canonical test application, requiring cooperation between peers for socially optimal outcomes. The Artificial Social Networks (ASN) produced are connected, cooperative and robust - possessing many of the disable properties of human friendship networks such as trust between friends (directly linked peers) and short paths linking everyone via a chain of friends. In addition to new application possibilities, SLACER could supply ASN to P2P applications that currently depend on human social networks thus transforming them into fully autonomous, self-managing systems.