Dependence and Relevance: A probabilistic view

TL;DR

This paper explores three probabilistic concepts related to independence and relevance, analyzing their relationships and implications for constructing similarity networks and Bayesian networks.

Contribution

It clarifies the relationships between probabilistic independence, relevance, and connectedness, enhancing understanding of probabilistic knowledge acquisition and network construction.

Findings

Established relationships between independence and relevance concepts.

Linked connectedness in Bayesian networks to relevance in probability.

Provided insights for constructing similarity networks from expert knowledge.

Abstract

We examine three probabilistic concepts related to the sentence "two variables have no bearing on each other". We explore the relationships between these three concepts and establish their relevance to the process of constructing similarity networks---a tool for acquiring probabilistic knowledge from human experts. We also establish a precise relationship between connectedness in Bayesian networks and relevance in probability.