Article Per Year (5 Year)
p-Index From 2021 - 2026
0.659
P-Index
This Author published in this journals
All Journal Journal of Natural Sciences and Mathematics Research Engineering, Mathematics and Computer Science Journal (EMACS) Journal of Fundamental Mathematics and Applications (JFMA) Indonesian Journal of Statistics and Its Applications
Sarini Abdullah
Department Of Mathematics, Faculty Of Mathematics And Natural Sciences, Universitas Indonesia, Indonesia
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Chemistry Civil Engineering, Building, Construction & Architecture Computer Science & IT Decision Sciences, Operations Research & Management Engineering Industrial & Manufacturing Engineering Materials Science & Nanotechnology Mathematics Physics Other

Published : 5 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 5 Documents
Search

 1 
EASY ENSEMMBLE WITH RANDOM FOREST TO HANDLE IMBALANCED DATA IN CLASSIFICATION Sarini Abdullah; GV Prasetyo
Journal of Fundamental Mathematics and Applications (JFMA) Vol 3, No 1 (2020)
Publisher : Diponegoro University

Show Abstract | Download Original | Original Source | Check in Google Scholar | Full PDF (1177.977 KB) | DOI: 10.14710/jfma.v3i1.7415

Abstract

Imbalanced data might cause some issues in problem definition level, algorithm level, and data level. Some of the methods have been developed to overcome this issue, one of state-of-the-art method is Easy Ensemble. Easy Ensemble was claimed can improve model performance to classify minority class, and overcome the deficiency of random under- sampling. In this paper we discussed the implementation of Easy Ensemble with Random Forest Classifiers to handle imbalance problem in credit scoring case. This combination method is implemented in two datasets which taken from data science competition website, finhacks.id and kaggle.com with class proportion within majority and minority is 70:30 and 94:6. The results showed that resampling with Easy Ensemble can improve Random Forest classifier performance upon minority class. Recall on minority class increased significantly after the resampling. Before resampling, the recall on minority class for the first dataset (finhacks.id) was 0.49, and increased to 0.82 after the resampling. Similar results were obtained for the second data set (kaggle.com), where the recall for the minority class was increased from just 0.14 to 0.73.
Bayesian Accelerated Failure Time Model for Risk Pregnancy Detection Dennis Alexander; Sarini Abdullah; Adam Fahsyah Nurzaman
Engineering, MAthematics and Computer Science Journal (EMACS) Vol. 5 No. 3 (2023): EMACS
Publisher : Bina Nusantara University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21512/emacsjournal.v5i3.10540

Abstract

Preeclampsia (PE) also known as a hypertension during third trisemester of pregnancy. PE, is one of the most feared complications of pregnancy because it can potentially become serious complications in the future, including mother and fetus’s death. The goal of this study is other than to have a bettter undestanding about risk factor in pregnancy by modelling the relationship between several factors and the time until deliveries under the PE condition. Data on 924 patients at obstetric and gynecology department in Jakarta were used in the analysis. Accelerated Failure Time (AFT) model was proposed to indentify some risk factors that influenced the condition. Model parameters were estimated using Bayesian method. Due to imbalance data, undersampling method will be used as a pre-procesing stage. Ratio between PE and non-PE data will be 60:40. Flat prior and posterior sample will be used using MCMC simulation with 12,000 iterations (including 2,000 iterations as a burnin stage) to get a convergen result. The iteration was repeated for 100 times so that the chosen data from undersampling was not error and biased. A consistent result for credible interval of the mean result was considered as the factors that affect PE condition consistently. From this study, there are two factors that have consistent Credible Interval result, Body Mass Index (BMI) and Mean Arterial Pressure (MAP).
ANALISIS TINGKAT KESEHATAN DAN EFISIENSI PERBANKAN TERHADAP PROFITABILITAS BANK MENGGUNAKAN REGRESI BERGANDA DAN ANOVA: Studi kasus pada tahun 2014 – 2017 Dita Anggun Lestari; Sarini Abdullah
Indonesian Journal of Statistics and Applications Vol 4 No 3 (2020)
Publisher : Statistics and Data Science Program Study, SSMI, IPB University, in collaboration with the Forum Pendidikan Tinggi Statistika Indonesia (FORSTAT) and the Ikatan Statistisi Indonesia (ISI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/ijsa.v4i3.538

Abstract

In this digital era, the competitiveness of small banks has decreased, and many bank consolidation phenomena have occurred. This study aims to examine the effect of bank soundness and efficiency on profitability in the face of competition and the current bank consolidation or merger phenomenon. Determination of variables refers to Bank Indonesia standards in measuring bank performance using the RGEC method approach consisting of the ratio of LDR, NIM, BOPO, NPL, CAR, and prime lending rate (SBDK), while bank profitability is represented by ROA. The research object is the bank category BUKU 1 - 4 which is supervised by OJK and listed as issuers on the Indonesia Stock Exchange during 2014 - 2017. The sampling technique used is purposive sampling so that from 102 banks 34 banks were obtained which were used as research objects. The data analysis technique used is multiple regression analysis and Anova comparison test. Based on the results of data testing, it is known that simultaneously and partially the ratios of LDR, NIM, BOPO, NPL, CAR, and SBDK have an effect on ROA. In comparison to the average BOPO, prime lending rate, and ROA variables, there are significant differences with bank categorization BUKU 1-4.
Early Preeclampsia Detection Using XGBoost-Cox Proportional Hazard Model Arya Wira Syahdwinata; Sarini Abdullah
Indonesian Journal of Statistics and Applications Vol 9 No 1 (2025)
Publisher : Statistics and Data Science Program Study, SSMI, IPB University, in collaboration with the Forum Pendidikan Tinggi Statistika Indonesia (FORSTAT) and the Ikatan Statistisi Indonesia (ISI)

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/ijsa.v9i1p33-45

Abstract

Various prognostic models based on survival analysis methods have been proposed to predict the risk of preeclampsia (PE). To develop a more accurate yet interpretable prediction approach, we utilized clinical data from pregnant women collected at a hospital in Jakarta and applied the XGBoost-Cox Proportional Hazard Model (XGB-Cox). This model integrates the predictive power of the XGBoost machine learning algorithm with the Cox Proportional Hazard (Cox-PH) model, which estimates the effect of covariates on event time. Our results show that the XGB-Cox model outperforms the traditional Cox-PH model based on four evaluation metrics: log-likelihood, log-rank test, concordance index (C-index), and Brier score. The XGB-Cox model achieved a higher C-index of 0.8908 compared to 0.7548 for Cox-PH, indicating improved risk discrimination. Kaplan-Meier curves suggest that XGB-Cox provides better separation across risk quartiles. While XGB-Cox generally yields lower Brier Scores, its performance declines at later gestational weeks. The Cox-PH model remains superior in interpretability, offering clear hazard ratios, while XGB-Cox enhances model fitness and still provides meaningful insights into feature importance. Additionally, sensitivity analysis underscores the need to carefully determine the proportion of censored data, as excessive censoring affects model stability. These findings suggest that XGB-Cox provides a robust predictive framework for early PE risk assessment, supporting its potential application in clinical decision-making for maternal healthcare.
A country-level k-Means Clustering of stress and coping strategies during the COVID-19 pandemic Stella Mediana Manalu; Sarini Abdullah
Journal of Natural Sciences and Mathematics Research Vol. 12 No. 1 (2026): June
Publisher : Faculty of Science and Technology, Universitas Islam Negeri Walisongo Semarang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.21580/jnsmr.v12i1.31506

Abstract

The COVID-19 pandemic has substantially increased psychological stress worldwide, making it important to understand how countries differ in their coping strategies. Previous cross-national studies have primarily relied on descriptive or regression-based analyses, which are limited in identifying latent national coping-stress profiles based on multiple coping behaviours. To address this limitation, this study applies K-Means clustering to identify groups of countries with similar stress and coping characteristics. The analysis was conducted using aggregated country-level data from the COVIDiSTRESS Global Survey. The final dataset consisted of 25 countries. For the clustering analysis, each country was represented using 19 features, consisting of the mean values of 16 coping 16 coping strategies and the proportions of respondents in the low, moderate, and high stress categories. The optimal number of clusters was determined using the elbow method dan supported by the silhouette coefficient, resulting in a four-cluster solution. Four distinct national coping-stress profiles were identified. Cluster 1 (Czech Republic, France, Hungary, Japan) exhibited balanced stress levels with moderate engagement in social dan recreational coping activities. Cluster 2 (Argentina, Bosnia and Herzegovina, Bulgaria, Croatia, Mexico, Poland, Turkey, United States) recorded the highest stress levels and was characterised by intensive social interaction and entertainment-based coping. Cluster 3 (nine European countries) exhibited the lowest stress levels and predominantly relied on institutional trust and government information, Cluster 4 (Indonesia, Kosovo, Portugal, Taiwan) showed moderate stress levels and was characterised by greater adherence to government-related information, entertainment-based coping, and the highest use of religious or spiritual coping. These findings demonstrate that K-Means clustering provides a useful approach for identifying cross-national stress-coping profiles that may not be evident through conventional statistical analyses. The identified clusters offer insights into variations in coping behaviours across countries and may support the development of more context-specific mental health interventions.
Co-Authors Adam Fahsyah Nurzaman Alexander Dennis Suwardi Arya Wira Syahdwinata Dita Anggun Lestari GV Prasetyo Rafif Hibatullah Stella Mediana Manalu Yekti Widyaningsih