Enhancing OLAP functionality using self-organizing neural networks

Authors:
Sohail Asghar;Damminda Alahakoon;Arthur Hsu
Affiliations:
School of Business Systems, Monash University, Clayton, Victoria 3800, Australia;School of Business Systems, Monash University, Clayton, Victoria 3800, Australia;Mechatronics Research Group, Department of Mechanical and Manufacturing Engineering, University of Melbourne, Victoria 3010, Australia
Venue:
Neural, Parallel & Scientific Computations - Special issue: Computing intelligence in management
Year:
2004

Citing 12
Cited 1

Growing cell structures—a self-organizing network for unsupervised and supervised learning

Neural Networks
An overview of data warehousing and OLAP technology

ACM SIGMOD Record
Advances in knowledge discovery and data mining

Advances in knowledge discovery and data mining
Self-organizing maps

Self-organizing maps
Towards on-line analytical mining in large databases

ACM SIGMOD Record
Data mining: concepts and techniques

Data mining: concepts and techniques
Data Mining and Uncertain Reasoning: An Integrated Approach

Data Mining and Uncertain Reasoning: An Integrated Approach
Database Technology for Decision Support Systems

Computer
Data Mining and Knowledge Discovery: Making Sense Out of Data

IEEE Expert: Intelligent Systems and Their Applications
Hierarchically Distributed Data Warehouse

HPC '00 Proceedings of the The Fourth International Conference on High-Performance Computing in the Asia-Pacific Region-Volume 2 - Volume 2
Improving OLAP Performance by Multidimensional Hierarchical Clustering

IDEAS '99 Proceedings of the 1999 International Symposium on Database Engineering & Applications
Dynamic self-organizing maps with controlled growth for knowledge discovery

IEEE Transactions on Neural Networks

Augmenting OLAP exploration with dynamic advanced analytics

Proceedings of the 13th International Conference on Extending Database Technology

Quantified Score

Hi-index	0.00

Visualization

Abstract

On-Line Analytical Processing (OLAP) has become a popular management decision-making tool due to its user-friendly visualization abilities. With this popularity, user's expectation and demands from existing OLAP systems have increased. The described work extends the capabilities of OLAP with additional functionality by using neural network technology. In addition to the usual visualization capabilities, this new technique provides the user with the opportunity to analyse clusters in the data at different levels of abstraction. The technique used for enhancing OLAP functionality is a model called the Growing Self-Organizing Map (GSOM). The GSOM has been developed as a more flexible data mining friendly feature mapping method over the traditional Self-Organizing Map (SOM). One of the major innovations with the GSOM is the possibility of generating feature maps of different levels of data abstraction using a parameter called the spread factor. This spread factor has been used to develop a hierarchical cluster generation and analysis technique called the Dynamic SOM Tree. These hierarchical clusters facilitate the OLAP user to gain insight and obtain prior knowledge of the data set before performing OLAP operations. In addition, the hierarchical clusters from a Dynamic SOM Tree are used to provide the OLAP user with the ability to visualize and select data clusters at different levels of abstraction for further detailed analysis.