Setiani, Dian (2020) IMPLEMENTASI METODE BACKPROPAGATION NEURAL NETWORK DALAM PENGENALAN KARAKTER TULISAN HIRAGANA JEPANG. Tugas Akhir thesis, University of Technology Yogyakarta.
|
Text
NASKAH PUBLIKASI 5160411024 Dian Agus Setiani.pdf Download (736kB) | Preview |
Abstract
In Japanese language learning process, mastering Hiragana Letters is the first step that must be passed. There are often obstacles in learning Hiragana letters because the Hiragana characters are different from the Latin letters that are often used by the public. Some learners are difficult to allocate their time if they have to attend a Japanese language course session. For this reason, the researcher builds a Hiragana character pattern recognition system using the Backpropagation Neural Network method and Zoning feature extraction. The input used is a handwritten image of Hiragana letters in .jpg format. The Hiragana letters are written by 6 people. Each writes 46 hiragana letters. In the training process, the training data used are 184 data while in the testing process, the test data used are 92 data. The achievement of the maximum level of accuracy in testing is 69.56% with training accuracy of 91.30%. Keywords: Hiragana letters, Backpropagation Neural Network, Zoning, Pattern Recognition
| Item Type: | Thesis (Skripsi, Tugas Akhir or Kerja Praktek) (Tugas Akhir) |
|---|---|
| Subjects: | T Technology > T Technology (General) |
| Divisions: | Fakultas Sains Dan Teknologi > S1 Informatika |
| Depositing User: | Kaprodi S1 Informatika UTY |
| Date Deposited: | 13 Oct 2020 04:06 |
| Last Modified: | 13 Oct 2020 04:06 |
| URI: | http://eprints.uty.ac.id/id/eprint/5799 |
Actions (login required)
 |
View Item |
