Home // ADAPTIVE 2012, The Fourth International Conference on Adaptive and Self-Adaptive Systems and Applications // View article

On The Adaptivity of Distributed Association Rule Mining Agents

Authors:
Adewale Opeoluwa Ogunde
Olusegun Folorunso
Adesina Simon Sodiya

Keywords: adaptive agents; distributed association rule mining; distributed databases; knowledge integration; mobile agents

Abstract:
Current and real association rule mining tasks can only be successfully done in a distributed setting where transaction data sites are mined dynamically and appropriately as they are updated. Mobile agents’ paradigms are now used to mine association rules in such circumstances. As these mobile agents travel in the distributed association rules mining environment, they are liable to unforeseen changes, circumstances and faults that may arise in these environments. Few researches had been carried out on the adaptivity of mobile agents, but the adaptivity of distributed association rule mining agents is yet to be explored. Therefore, this work examines an adaptive architectural framework that mines association rules across multiple data sites, and more importantly the architecture adapts to changes in the updated database and the mining environment giving special considerations to the incremental database. This system was made adaptive both at the algorithm level and the mining agent level. Adaptation at the mobile agent level uses sensors to sense environmental changes, creates a percept of the environment and sends it to the adapter which adapts to the environmental changes by dynamically changing the goals of the mining agents or maintaining the original goals. The system promises to efficiently generate new and up-to-date rules while also adapting to faults and other unforeseen circumstances in the distributed association rules mining environment without the usual user’s interference. The model presented here provided the background ideas needed for the development of adaptive distributed association rule mining agents.

Pages: 69 to 74

Copyright: Copyright (c) IARIA, 2012

Publication date: July 22, 2012

Published in: conference

ISSN: 2308-4146

ISBN: 978-1-61208-219-6

Location: Nice, France

Dates: from July 22, 2012 to July 27, 2012