Map-reduce implementation of belief combination rules

Frédéric Dambreville (1, 2)

1 Lab-STICC_ENSTAB_CID_SFIIS
2 Pôle STIC_REMS

This paper presents a generic and versatile approach for implementing combining rules on preprocessed belief functions, issuing from a large population of information sources. In this paper, we address two issues, which are the intrinsic complexity of the rules processing, and the possible large amount of requested combinations. We present a fully distributed approach, based on a map-reduce (Spark) implementation.

Mots clés: Belief functions; combination rules; statistic; map-reduce

Domaines: Intelligence artificielle [cs.AI]; Complexité [cs.CC]; Réseau de neurones [cs.NE]; Théorie de l'information [cs.IT]; Systèmes et contrôle [cs.SY]

Frédéric Dambreville. Map-reduce implementation of belief combination rules. DATA 2016, Jun 2016, Lisbonne, Portugal. ⟨hal-01331654⟩