Article Per Year (5 Year)
p-Index From 2020 - 2025
0.23
P-Index
This Author published in this journals
All Journal Bulletin of Social Informatics Theory and Application
Kuswanto , Djoko
Unknown Affiliation
Computer Science & IT Social Sciences

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

Found 1 Documents
Search

 1 
Data mining for forecasting community mobility denpasar city with long short-term memory method Setiawan , I Wayan Agus Hery; Triandini, Evi; Suniantara , I Ketut Putu; Kuswanto , Djoko
Bulletin of Social Informatics Theory and Application Vol. 8 No. 2 (2024)
Publisher : Association for Scientific Computing Electrical and Engineering

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31763/businta.v8i2.670

Abstract

Denpasar City has a high potential for community mobility, this is supported by many public facilities. High and highly volatile human mobility causes the transmission of the COVID-19 virus to spread very quickly, so forecasting is needed to find out a picture of future community mobility using data mining techniques. Data mining is the process of solving problems by analyzing data that already exists in the database. Denpasar City community mobility data for the period September 1, 2021 – October 31, 2021 show that most of the high mobility is in the junior high school sector. The Long Short-Term Memory method was chosen as a method that can assist in forecasting community mobility. Long Short-Term Memory has the advantage of dealing with missing gradient problems and can be used on all types of data patterns, whether trend, cyclical, seasonal, or horizontal patterns. Hyperparameter tests were carried out including LSTM_units representing the number of Long Short-Term Memory units in each layer, Dropout, and Optimizer to obtain the optimal prediction method. this combination yields a total of 45 methods. The best hyperparameter obtained is at LSTM_units of 128, Dropout of 0.1, and Optimizer is Adam. The results obtained with this hyperparameter are the Root Mean Square Error (RMSE) value of 971,438687. This method results in forecasting the mobility of the people of Denpasar City from November 1, 2021 to November 7, 2021, reaching 9.550 total checkins which is close to the actual value of 10.219
Co-Authors Evi Triandini Setiawan , I Wayan Agus Hery Suniantara , I Ketut Putu