Consid{\'e}rant la d{\'e}pendance dans la th{\'e}orie des fonctions de croyance

TL;DR

This paper introduces a method to assess source independence in belief function theory, enabling the selection of suitable combination rules based on source dependence or independence, thereby improving belief aggregation accuracy.

Contribution

It proposes a new measure of source independence, including positive and negative dependence, to inform the choice of combination rules in belief function theory.

Findings

A measure of source independence is developed.

Incorporating dependence measures improves belief combination.

Guidelines for selecting combination rules based on source dependence.

Abstract

In this paper, we propose to learn sources independence in order to choose the appropriate type of combination rules when aggregating their beliefs. Some combination rules are used with the assumption of their sources independence whereas others combine beliefs of dependent sources. Therefore, the choice of the combination rule depends on the independence of sources involved in the combination. In this paper, we propose also a measure of independence, positive and negative dependence to integrate in mass functions before the combinaision with the independence assumption.