Hanifah Az-Zahra, . (2026) RANCANG BANGUN APLIKASI PREDIKSI HARGA CABAI MENGGUNAKAN LSTM. Skripsi thesis, Universitas Pembangunan Nasional Veteran Jakarta.
![[img]](http://repository.upnvj.ac.id/style/images/fileicons/text.png) |
Text
ABSTRAK.pdf Download (204kB) |
|
Text
AWAL.pdf Download (758kB) |
|
Text
BAB 1.pdf Restricted to Repository UPNVJ Only Download (244kB) |
|
Text
BAB 2.pdf Restricted to Repository UPNVJ Only Download (575kB) |
|
Text
BAB 3.pdf Restricted to Repository UPNVJ Only Download (604kB) |
|
Text
BAB 4.pdf Restricted to Repository UPNVJ Only Download (2MB) |
|
Text
BAB 5.pdf Download (233kB) |
|
Text
DAFTAR PUSTAKA.pdf Download (215kB) |
|
Text
RIWAYAT HIDUP.pdf Restricted to Repository UPNVJ Only Download (170kB) |
|
Text
LAMPIRAN.pdf Restricted to Repository UPNVJ Only Download (743kB) |
|
Text
HASIL PLAGIARISME.pdf Restricted to Repository staff only Download (10MB) |
|
Text
ARTIKEL KI.pdf Restricted to Repository staff only Download (442kB) |
Abstract
Chili is one of the most widely consumed horticultural commodities in Indonesia; however, its price is highly volatile. These price fluctuations can make it difficult for consumers, traders, and Micro, Small, and Medium Enterprises (MSMEs) to plan their purchasing and production activities. This study aims to develop an Android-based chili price prediction application covering 34 provinces in Indonesia by utilizing a Long Short-Term Memory (LSTM) model. Historical chili price data were preprocessed before being used to train the LSTM model. The trained model was then integrated into an Android application developed using Flutter, with FastAPI serving as the backend, to generate price predictions based on parameters selected by users. Model performance was evaluated using Mean Absolute Error (MAE), Root Mean Squared Error (RMSE), and Mean Absolute Percentage Error (MAPE), and further validated using Time Series Cross-Validation. The application was tested using Black Box Testing and User Acceptance Testing (UAT) to assess its functionality and level of user acceptance. The results show that the application was successfully developed and is capable of presenting chili price predictions for periods of 1, 7, 14, and 30 days in the form of graphs, prediction trend summaries, and prediction tables, complemented by an online store recommendation feature. The LSTM model achieved a global MAPE of 5.63%, a global RMSE of 4049, and a global MAE of 2830 on the final test data. Validation using Time Series Cross-Validation produced an average MAPE of 6.46% with a standard deviation of 2.72%, an average RMSE of 4330, and an average MAE of 2859. The UAT results showed a user acceptance rate of 90.20%, categorized as Very Good, while Black Box Testing confirmed that all application features functioned according to the specified functional requirements.
| Item Type: | Thesis (Skripsi) |
|---|---|
| Additional Information: | [No.Panggil:2210511046] [Pembimbing: Indra Permana Solihin] [Penguji 1: Jayanta] [Penguji 2: Kharisma Wiati Gusti] |
| Uncontrolled Keywords: | chili price prediction, Long Short-Term Memory, Android application |
| Subjects: | Q Science > Q Science (General) |
| Divisions: | Fakultas Ilmu Komputer > Program Studi Informatika (S1) |
| Depositing User: | HANIFAH AZ-ZAHRA |
| Date Deposited: | 31 Jul 2026 01:34 |
| Last Modified: | 31 Jul 2026 01:34 |
| URI: | http://repository.upnvj.ac.id/id/eprint/52249 |
Actions (login required)
 |
View Item |