Skip to main content
WorldCist'18 - 6th World Conference on Information Systems and Technologies

Full Program »

Adapting a Multi-SOM clustering algorithm to large banking data

It the recent years, Big Data (BD) has attracted researchers in many domains as a new concept providing opportunities to improve research applications including business, science, engineering. Big Data Analytics is becoming a practice that many researchers adopt to construct valuable information from BD. This paper presents the BD technologies and how BD is useful in Cluster Analysis. Then, a clustering approach named multi-SOM is studied. In doing so, a banking dataset is analyzed integrating R statistical tool with BD technologies that include Hadoop Distributed File System, HBase and Map Reduce. Hence, we aim to decrease the time execution of multi-SOM clustering method in determining the number of clusters using R and Hadoop. Results show the performance of integrating R and Hadoop to handle big data using multi-SOM clustering algorithm and to overcome the weaknesses of R.

Imen Khanchouch
ISG Tunis
Tunisia

Mohamed Limam
Dhofar University
Oman

 

Powered by OpenConf®
Copyright ©2002-2017 Zakon Group LLC