Clustering belief functions based on attracting and conflicting metalevel evidence

TL;DR

This paper introduces a clustering method for belief functions that leverages attracting and conflicting metalevel evidence to separate belief functions into independent subsets, aiding in better handling of multiple events.

Contribution

It presents a novel clustering approach utilizing both internal conflict and external attraction evidence to improve belief function separation.

Findings

Effective separation of belief functions into meaningful clusters

Utilizes both internal conflict and external attraction evidence

Applicable to complex multi-event scenarios

Abstract

In this paper we develop a method for clustering belief functions based on attracting and conflicting metalevel evidence. Such clustering is done when the belief functions concern multiple events, and all belief functions are mixed up. The clustering process is used as the means for separating the belief functions into subsets that should be handled independently. While the conflicting metalevel evidence is generated internally from pairwise conflicts of all belief functions, the attracting metalevel evidence is assumed given by some external source.