Article Per Year (5 Year)
p-Index From 2020 - 2025
0
P-Index
This Author published in this journals
All Journal Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer
Davia Werdiastu
Fakultas Ilmu Komputer, Universitas Brawijaya
Computer Science & IT Control & Systems Engineering Education Electrical & Electronics Engineering Engineering

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Estimasi Hasil Produksi Benih Berdasarkan Karakteristik Tanaman Kenaf Menggunakan Metode Backpropagation (Studi Kasus: Balai Tanaman Pemanis dan Serat Kota Malang) Davia Werdiastu; Dian Eka Ratnawati; Bayu Rahayudi
Jurnal Pengembangan Teknologi Informasi dan Ilmu Komputer Vol 2 No 11 (2018): November 2018
Publisher : Fakultas Ilmu Komputer (FILKOM), Universitas Brawijaya

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1107.432 KB)

Abstract

Kenaf plants have many benefits. However, currently it has limitation production of Kenaf palnts. According to Research and Development Agency, Malang City stated that Kenaf seed production was just about 0.3-0.5 tons/ ha, while farmers' need for superior seed of Kenaf plants was about 0.7-1.0 tons/ ha. Balai Penelitian Tanaman Pemanis dan Serat (BALITTAS) Malang city was directly elected to carry out certification of seed consist of field inspection, laboratory test, and labeling. Seed certification aimed to ensure seeds quality. For seeds certification, BALITTAS has difficult to estimate resulted seeds. This estimate was required to prepare certification requirements such as laboratory equipment, yarn, gunny sack, and workers. This can be solved by built an estimation system using backpropagation algorithm. The number of neurons in the input layer was 4 inputs ie the number of seeds production was the age of flower I, bottom diameter, the weight of 10 plants seeds, and the number of mature capsules, and produced 1 output as resulted seeds. The calculation process starts from initialled initial weight with nguyen-widrow, feedforward and backpropagation, then update weight and bias. Test result showed the best mean of MAPE value was 0,938% with 90% testing test scenario, 10% test data, 5 neurons in hidden layer, learning rate 0,3, maximum 1% MAPE and maximum limit of iteration 5000.
Co-Authors Bayu Rahayudi Dian Eka Ratnawati