Prediksi Hasil Panen Benih Tanaman Kenaf Menggunakan Metode Support Vector Regression (SVR) Pada Balai Penelitian Tanaman Pemanis dan Serat (Balittas)

Robih Dini (Fakultas Ilmu Komputer, Universitas Brawijaya)
Budi Darma Setiawan (Fakultas Ilmu Komputer, Universitas Brawijaya)
Candra Dewi (Fakultas Ilmu Komputer, Universitas Brawijaya)

Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Vol 2 No 12 (2018): Desember 2018

Kenaf (Hibiscus cannabinus L.) is a fiber plant that has many benefits. Kenaf is grown by seed so it is necessary to handle the seeds in order to ensure the quality of the seed is not decreased so as to increase the productivity of the kenaf. Balai PenelitianTanaman Pemanis dan Serat (Balittas) as the producer of the seeds has constraint to predict the yields of kenaf seed for the proper handling preparation of kenaf seeds. Therefore in this research proposed regression method using Support Vector Regression (SVR) by using Radial Basis Function (RBF). Hopefully this research can help Balittas to prepare the handling of the harvested of kenaf seeds properly. The research used 100 data about the characteristics of kenaf plants measured from the beginning of planting until the time of harvest. From the testing results that have been done, the result of prediction show the error value using Mean Absolute Percentage Error 3,5371% by using the best SVR parameters value which is cLR = 0,01, σ = 0,25, ε = 1 x 10-7, C = 0,5, λ = 0,6, and the number of iterations = 25000.