Innovation of Large-scale Consensus Support System based on Agent Technologies:A Large-scale Social Experiment using the Automated Facilitation Agent

Abstract

Crowd-scale discussion platforms are receiving great attention as potential next-generation methods for democratic citizen platforms. Such platforms require support functions that can efficiently achieve a consensus, reasonably integrate ideas, and discourage flaming. We are developing several crowd-scale discussion platforms and conducting social experiments with private citizens. The aim of this paper is to present an implementation of the automated facilitation agent that facilitate online crowd discussion and demonstrate the result of our large scale social experiment with Nagoya local government. Our automated facilitation agent extracts discussion structures from text discussions, analyzes them, and posts facilitation messages. The experiment results demonstrate that the automated facilitation agent worked quite well.