IMPLEMENTASI SUPPORT VECTOR REGRESSION UNTUK PREDIKSI CUACA PERJAM PADA DATA NYC HOURLY TEMPERATURE
| Main Author: | Akmal, Nurfaizin |
|---|---|
| Format: | Thesis NonPeerReviewed Book |
| Bahasa: | eng |
| Terbitan: |
, 2019
|
| Subjects: | |
| Online Access: |
http://eprints.umm.ac.id/57632/1/PENDAHULUAN.pdf http://eprints.umm.ac.id/57632/2/BAB%20I.pdf http://eprints.umm.ac.id/57632/3/BAB%20II.pdf http://eprints.umm.ac.id/57632/4/BAB%20III.pdf http://eprints.umm.ac.id/57632/5/BAB%20IV.pdf http://eprints.umm.ac.id/57632/6/BAB%20V.pdf http://eprints.umm.ac.id/57632/7/LAMPIRAN.pdf http://eprints.umm.ac.id/57632/ |
Daftar Isi:
- Knowing kondition weather temperature is very important in various activity. But sometimes the level of prediction still not meet expectation. To get a good result research should be supported by appropiate data and methods. The main porpuse of this research is to develop prediction model using support vector regression with three kernels ie linier, polynomial, radial basis function for weather temperature prediction. Each kernels have a parameter value must be filled, search the optimal parameter value using grid search method. The testing did it with dividing the data set into data train and data tes using k-fold cross validation. From a series of the tests have been done the best result is using radial basis function kernel with parameter value 3, get the averange of root mean square error is 10.153 and averange root relative square error is 0.005 while for the averange koefisien korelasi is 4.24