Hybrid Genetic Algorithm On Predicting Next Rainfall Distribution
In meteorology, the small changes in the initial condition of the atmosphere will lead to big changes in future weather classification. It is indeed considerably sensitive as small changes in the state of the weather can cause big differences in the future weather. Meteorologists who are experts in...
Saved in:
| Main Author: | |
|---|---|
| Format: | Thesis |
| Language: | English English |
| Published: |
2017
|
| Subjects: | |
| Online Access: | http://eprints.utem.edu.my/id/eprint/20726/1/Hybrid%20Genetic%20Algorithm%20On%20Predicting%20Next%20Rainfall%20Distribution%20-%20Rasuan%20Mohd%20Zain%20-%2024%20Pages.pdf http://eprints.utem.edu.my/id/eprint/20726/2/Hybrid%20Genetic%20Algorithm%20On%20Predicting%20Next%20Rainfall%20Distribution.pdf |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| Summary: | In meteorology, the small changes in the initial condition of the atmosphere will lead to big changes in future weather classification. It is indeed considerably sensitive as small changes in the state of the weather can cause big differences in the future weather. Meteorologists who are experts in understanding weather, gather information which can hopefully help humans cope with weather unpredictability. In doing weather classification, the expert must have broad knowledge about the weather and also the variability related with it and they should be able to deal with any chaotic atmosphere. Based on that, the weather classification always require critical analysis and it is not an easy task to achieve complete accuracy because of chaotic weather data. The existing conventional technique also could not deal with chaotic weather data effectively. Hence, the performance of existing techniques are not evaluated thoroughly. It is found in related literatures, in particular soft computing techniques; Artificial Intelligent (GA) has provided an alternative solution in doing weather classification. The literature has emphasized the importance of hybrid techniques which have been widely used among researchers in many areas such as patterns recognition, making predictions, medical diagnosis and such other applications. A hybrid technique is a potentially powerful tool that may enable us to address and solve problems that are just too complex for conventional approaches (Jackson, 1999). |
|---|