Daffa Bagus Maulana, . (2026) RANCANG BANGUN APLIKASI WEB PEMBELAJARAN KOSAKATA JEPANG BERBASIS MOBILENETV2 UNTUK KLASIFIKASI OBJEK DENGAN PENERAPAN GAMIFIKASI. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (377kB) |
|
Text
AWAL.pdf Download (1MB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (519kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (3MB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (3MB) |
|
Text
BAB 5.pdf Download (499kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (437kB) |
|
Text
DAFTAR RIWAYAT HIDUP.pdf Restricted to Repository staff only Download (91kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (1MB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (22MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (552kB) |
Abstract
Indonesia ranks second among countries with the largest number of Japanese language learners, yet vocabulary learning still relies heavily on text-based media that provide limited contextual and interactive experiences. This study aims to design and develop GoiHunt as a Japanese vocabulary learning medium based on object classification and gamification. The application was developed using an Agile approach through planning, design, development, testing, and deployment stages. The classification model was built using transfer learning with ImageNet pretrained weights through two experimental stages. The first stage compared three augmentation scenarios on MobileNetV2: no augmentation, light augmentation, and heavy augmentation. The no augmentation scenario achieved a test accuracy of 96.06% and a test loss of 0.0999. The second stage compared MobileNetV2 and MobileNetV3Large. MobileNetV3Large achieved the highest test accuracy of 98.79%, while MobileNetV2 achieved 96.06%. However, MobileNetV2 had a smaller model size, 8.66 MB compared with 11.47 MB, and was therefore selected as the final model to support local browser-based inference. The model was converted to TensorFlow.js and integrated into a React.js web application. Black-box testing across 44 scenarios produced a 100% success rate. User Acceptance Testing involving 40 respondents resulted in an average acceptance score of 89%, categorized as Very Good. The learning evaluation also indicated improved cognitive outcomes. The results show that GoiHunt is suitable as a visual, contextual, interactive, and online-accessible supporting medium for Japanese vocabulary learning.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil: 2210511063] [Pembimbing 1: Neny Rosmawarni] [Pembimbing 2: I Wayan Rangga Pinastawa] [Penguji 1: Muhammad Adrezo] [Penguji 2: Ichsan Mardani] |
| Uncontrolled Keywords: | MobileNetV2, Object Classification, Gamification, Japanese Vocabulary, Web Application |
| Subjects: | Q Science > QA Mathematics > QA75 Electronic computers. Computer science Q Science > QA Mathematics > QA76 Computer software |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | DAFFA BAGUS MAULANA |
| Date Deposited: | 03 Sep 2026 03:20 |
| Last Modified: | 03 Sep 2026 03:20 |
| URI: | http://repository.upnvj.ac.id/id/eprint/50942 |
Actions (login required)
 |
View Item |