Social Decision Making with Multi-Relational Networks and Grammar-Based Particle Swarms

TL;DR

This paper introduces a novel framework combining multi-relational networks and grammar-based particle swarms to enhance social decision support systems, enabling scalable aggregation of diverse individual inputs into collective decisions.

Contribution

It presents a new ontology and algorithmic approach that supports complex vote aggregation, accounting for expertise and representation across various problem domains.

Findings

Supports aggregation of decisions from millions of individuals.

Accommodates diverse problem spaces and vote types.

Effectively integrates individual expertise into collective outcomes.

Abstract

Social decision support systems are able to aggregate the local perspectives of a diverse group of individuals into a global social decision. This paper presents a multi-relational network ontology and grammar-based particle swarm algorithm capable of aggregating the decisions of millions of individuals. This framework supports a diverse problem space and a broad range of vote aggregation algorithms. These algorithms account for individual expertise and representation across different domains of the group problem space. Individuals are able to pose and categorize problems, generate potential solutions, choose trusted representatives, and vote for particular solutions. Ultimately, via a social decision making algorithm, the system aggregates all the individual votes into a single collective decision.