Setyowati, Silfiana Lis
Unknown Affiliation

Published : 3 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 3 Documents
Search

Comparison of Ordinal Logistic Regression and Geographically Weighted Ordinal Logistic Regression (GWOLR) in Predicting Stunting Prevalence among Indonesian Toddlers Setyowati, Silfiana Lis; Indahwati; Fitrianto, Anwar; Erfiani; Aliu, Muftih Alwi
Sainmatika: Jurnal Ilmiah Matematika dan Ilmu Pengetahuan Alam Vol. 21 No. 2 (2024): Sainmatika : Jurnal Ilmiah Matematika dan Ilmu Pengetahuan Alam
Publisher : Universitas PGRI Palembang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.31851/sainmatika.v21i2.15416

Abstract

Ordinal logistic regression is a type of logistic regression used for response variables with an ordinal scale, containing two or more categories with levels between them. This method is an extension of logistic regression where the observed response variable is ordinal with a clear order. It addresses spatial effects that can cause variance heterogeneity and improve parameter estimation accuracy compared to logistic regression. Geographically Weighted Regression (GWR) is a statistical analysis technique designed to account for spatial heterogeneity. GWOLR is an extension of OLS and GWR models that incorporates spatial elements into regression with categorical variables. This study compares the effectiveness of OLR and GWOLR in analyzing stunting prevalence in toddlers. Comparing OLR and GWOLR can help assess the spatial impact on stunting prevalence. This analysis could reveal that certain regions have a higher tendency for stunting prevalence, while others might have lower tendencies, thus helping in understanding regional disparities. Toddler height is a key indicator of health and nutrition in early growth. The prevalence of stunting for toddlers, according to WHO, is categorized into four levels: low, moderate, high, and very high. The Ordinal Logistic Regression model is better suited for modeling toddler stunting prevalence in Indonesia than the GWORL model. The Ordinal Logistic Regression model and the GWOLR both have a classification accuracy of 85.7%, but the OLR model has a lower AIC value. The GWOLR model is not suitable for analyzing stunting prevalence among Indonesian toddlers due to the lack of spatial variability in the data. The Breusch-Pagan test results indicate that there is no spatial heterogeneity in the data on stunting prevalence among Indonesian toddlers, as the p-value is less than the significance level of 0.05. The prevalence of undernourished toddlers is the main factor influencing stunting among Indonesian toddlers.
Comparison of Seasonal ARIMA and Support Vector Machine Forecasting Method for International Arrival in Lombok MY, Hadyanti Utami; Setyowati, Silfiana Lis; Notodiputro, Khairil Anwar; Angraini, Yenni; Mualifah, Laily Nissa Atul
Jambura Journal of Mathematics Vol 6, No 2: August 2024
Publisher : Department of Mathematics, Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjom.v6i2.26478

Abstract

Seasonal Autoregressive Integrated Moving Average is a statistical model designed to analyze and forecast data with that shows seasonal patterns and trends. Support Vector Machine (SVM) is a machine learning-based technique that can be used to forecast time series data. SVM uses the kernel tricks to overcome non-linearity problems, whereas The SARIMA model is well-suited for data that exhibit seasonal fluctuations that repeat over time. Lombok International Airport is the main gateway to West Nusa Tenggara and has become a symbol of tourism growth in the region. Time series analysis is a very useful tool in determining patterns and forecasting the number of international arrivals at Lombok International Airport within a certain period. This study aims to compare the SARIMA model and SVM which can read non-linear patterns in the number of international arrivals at Lombok International Airport. After obtaining the SARIMA and SVM models, the two models are evaluated using test data based on the smallest RMSE value. The SVM model with a linear kernel trick provides the smallest RMSE when compared to SARIMA with SVM RMSE is 238,655. While the best model in Seasonal ARIMA is SARIMA (3,1,0)(1,0,0)12, the forecasting results show SARIMA is better in the forecasting process for the next 10 months.
Optimizing Random Forest Parameters with Hyperparameter Tuning for Classifying School-Age KIP Eligibility in West Java Setyowati, Silfiana Lis; Qalbi, Asyifah; Aristawidya, Rafika; Sartono, Bagus; Firdawanti, Aulia Rizki
Jambura Journal of Mathematics Vol 7, No 1: February 2025
Publisher : Department of Mathematics, Universitas Negeri Gorontalo

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.37905/jjom.v7i1.28736

Abstract

Random Forest is an ensemble learning algorithm that combines multiple decision trees to generate a more stable and accurate classification model. This study aims to optimize Random Forest parameters for classifying school-age students' eligibility for the Kartu Indonesia Pintar (KIP) in West Java, based on economic factors. The research uses secondary data from the 2023 National Socio-Economic Survey (SUSENAS) of West Java, with a sample size of 13,044 individuals. To address class imbalance, Synthetic Minority Oversampling Technique (SMOTE) is applied. Hyperparameter tuning through grid search identifies the optimal combination of parameters, including the number of trees (ntree), random variables per split (mtry), and terminal node size (node_size). Model performance is evaluated using balanced accuracy, sensitivity, and specificity. Results indicate that the optimal parameters (mtry = 5, ntree = 674, node_size = 26) yield a balanced accuracy of 65.47%. Significant variables include PKH status, floor area of the house, source of drinking water, and building material type. The model accurately identifies students in need of educational assistance. In conclusion, optimizing Random Forest parameters improves the accuracy of KIP eligibility classification, supporting educational equity policies in West Java. These findings provide a foundation for developing more effective beneficiary selection systems for educational aid.