Journal of New Democratic Methods: An Introduction

TL;DR

This paper introduces the Journal of New Democratic Methods, a novel academic journal that employs machine learning to democratize article submission and review processes, aiming to enhance representation and inclusivity.

Contribution

It presents a prototype of a democratic journal utilizing machine learning to evaluate reviewer accuracy and discusses broader implications for democratic systems.

Findings

Prototype of a democratic journal implemented

Machine learning effectively identifies representative reviewers

Potential for broader democratic applications discussed

Abstract

This paper describes a new breed of academic journals that use statistical machine learning techniques to make them more democratic. In particular, not only can anyone submit an article, but anyone can also become a reviewer. Machine learning is used to decide which reviewers accurately represent the views of the journal's readers and thus deserve to have their opinions carry more weight. The paper concentrates on describing a specific experimental prototype of a democratic journal called the Journal of New Democratic Methods (JNDM). The paper also mentions the wider implications that machine learning and the techniques used in the JNDM may have for representative democracy in general.