IMPLEMENTASI LONG SHORT TERM MEMORY (LSTM) UNTUK PREDIKSI HARGA CABAI DI PROVINSI JAWA TIMUR

Fata Nabil Fikri, NIM.: 19106050034 (2023) IMPLEMENTASI LONG SHORT TERM MEMORY (LSTM) UNTUK PREDIKSI HARGA CABAI DI PROVINSI JAWA TIMUR. Skripsi thesis, UIN SUNAN KALIJAGA YOGYAKARTA.

Preview	Text (IMPLEMENTASI LONG SHORT TERM MEMORY (LSTM) UNTUK PREDIKSI HARGA CABAI DI PROVINSI JAWA TIMUR) 19106050034_BAB-I_IV-atau-V_DAFTAR-PUSTAKA.pdf - Published Version Download (2MB) | Preview
	Text (IMPLEMENTASI LONG SHORT TERM MEMORY (LSTM) UNTUK PREDIKSI HARGA CABAI DI PROVINSI JAWA TIMUR) 19106050034_BAB-II_sampai_SEBELUM-BAB-TERAKHIR.pdf Restricted to Registered users only Download (6MB) | Request a copy

Abstract

In Indonesia, groceries have fluctuating or unstable prices over time and each region has different prices from other regions. One of groceries commodity that often encounter price instability is chili commodities. For this reason, the prediction for chili prices can be taken to estimate the next price so that the appropriate strategy can be taken. In this era of computing evolution, prediction can be carried out using machines through a machine learning process, especially the use of the LSTM artificial neural network for chili price predictions. In this research, experimental testing was carried out regarding hyperparameters and the structure of the LSTM network itself which was used to predict prices for two types of chilies in East Java Province, namely red chilies and rawit chilies. The results of this research show that the best hyperparameter configuration and network structure are the same for each type of chili price data tested. Red chili price data shows the best RMSE average value of 1751.690, while rawit chili price data shows the best RMSE average value of 1888.741.

Share this knowledge with your friends :

Actions (login required)

View Item