Data mining for decision making in water resources
conference contribution
posted on 2017-12-06, 00:00authored bySaleh Wasimi, Mohammad Mondal
The volume of water resources data in the world today is increasing at a phenomenal rate. Typically, different aspects of water resources are studied separately and the results combined using heuristics, hierarchy, categorization, zoning or abstract principles for a holistic outlook. The alternative is a simultaneous study of all aspects, which has some advantages. Due to need to handle different types and enormous volume of data in this approach, data mining techniques are required. Data mining looks for patterns, associations and links that lie hidden in the vast amount of data collected on various aspects of a water resources system. The common data mining strategies are: Classification, Clustering, Association, and Estimation. The common data mining tools can be classified under five broad groups of algorithms, which are: function estimation-based, lazy learning-based, meta learning- based, probability-based, and tree-based. These are illustrated with the IRIS data available at UCI data repository.
Funding
Category 1 - Australian Competitive Grants (this includes ARC, NHMRC)
History
Start Page
583
End Page
590
Number of Pages
8
Start Date
2007-01-01
ISBN-13
9843000003036
Location
Dhaka, Bangladesh
Publisher
Credence Printing
Place of Publication
Dhaka, Bangladesh
Peer Reviewed
Yes
Open Access
No
External Author Affiliations
Faculty of Business and Informatics; Institute of Water and Flood Management;
Era Eligible
Yes
Name of Conference
International Conference on Water and Flood Management