1. Title, Journal and Authors
- Title : A new rainfall forecasting modeling using the CAPSO algorithm and an artifical neural network
- Journal : Neural Comput & Applic
- Authors : Zahra Beheshti1, Morteza Firouzi2, Siti Mariyam Shamsuddin1, Masoumeh Zibarzani3, Zulkifli Yusop2
1 : UTM Big Data Centre, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
2 : Centre for Environmental Sustainability and Water Security (IPASA), Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
3 : Department of Information System, Faculty of Computing, Universiti Teknologi Malaysia, 81310 Skudai, Johor, Malaysia
2. Summary
- This paper demonstrated that the developed hybrid models to forecast rainfall had better performance than existing single model
based on artificial neural network (ANN)
- The hybrid models were trained by three meta-heuristic algorithms, CAPSO, GSA and ICA differently from the existing ANN model.
* CAPSO : Centripetal accelerated particle swarm optimization
* GSA : Gravitational search algorithm
* ICA : Imperialist competitive algorithm
- The proposed methods integrated the accuracu and structure of ANN.
- And, the hybrid learning method of ANN with the CAPSO algorithm provided better performances on testing data
3. Application to research
- These proposed training methods with several algorithms can provide detailed insight information into the processes
that are to overcome of ANN's weakness and to improve the machine learning model
4. contact
Heewon Jeong (Ph.D. program)
Environmental Systems Engineering Lab.
School of Environmental Science % Engineering
Gwangju Institute of Science and Technology
1 Oryong-dong Buk-gu Gwangju, 500-712, Korea
Phone : +82-10-8734-8657
Email : gua01114@gist.ac.kr