[Td18a] Relevant Filtering in a Distributed Content-based Publish/Subscribe System

Chapitres de Livre : Titre du livre: "NoSQL Data Models - Trends and Challenges", July 2018, John Wiley & Sons, pp. 193--226, (isbn: 978-1-78630-364-6)

Mots clés: Pub/Sub, NoSQL, TDV, Streaming

Résumé: This chapter focuses on the way to enhance relevance of filtering and to integrate such a process in two different implementations: a centric-based version and a distributed version in a NoSQL environment. Our contributions in this paper are: – definitions for novelty and diversity in this particular context, along with a pro- posal for a weighting score (Term Discrimination Values - TDV) adapted to the characteristics of items and subscriptions; – an efficient filtering algorithm for real-time Pub/Sub systems based on novelty and diversity which exploits redundancy between subscriptions’ history. Two optimized implementations are proposed in centralized and distributed contexts; – a validation which highlights the complementarity of novelty and diversity both in centralized and NoSQL environments; – enhancement of TDV [WIL 85] computation by proposing incremental versions in a distributed environment.


@inbook {
title="{NoSQL Data Models - Trends and Challenges}",
chapter="{Relevant Filtering in a Distributed Content-based Publish/Subscribe System}",
author="N. Travers and C. du Mouza",
publisher="John Wiley & Sons",