Garuda - Garba Rujukan Digital

PERFORMA KLASIFIKASI DATA TIDAK SEIMBANG DENGAN PENDEKATAN MACHINE LEARNING (STUDI KASUS: DIABETES INDIAN PIMA) Masjidil Aqsha; Nurtiti Sunusi
Jurnal Matematika UNAND Vol 12, No 2 (2023)
Publisher : Departemen Matematika dan Sains Data FMIPA Universitas Andalas Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/jmua.12.2.176-193.2023

Diabetes merupakan suatu penyakit atau gangguan metabolisme kronis dengan multi etiologi yang ditandai dengan tingginya kadar gula darah disertai dengan gangguan metabolisme karbohidrat, lipid, dan protein sebagai akibat insufisiensi fungsi insulin. Faktor risiko diabetes berhubungan dengan status diabetes sesorang. Berbagai pendekatan machine learning menjadi alternatif dalam memprediksi status diabetes. Namun, dalam banyak kasus, data yang tersedia tidak cukup seimbang dalam kelas datanya. Adanya ketidakseimbangan data dapat menyebabkan hasil prediksi menjadi tidak akurat. Tujuan penelitian dalam paper ini adalah untuk mengatasi masalah ketidakseimbangan data dan membandingkan kinerja model dalam memprediksi status diabetes. Secara umum, metode seperti Synthetic Minority Over-sampling Technique (SMOTE) dan Adaptive Synthetic (ADASYN) dapat digunakan untuk menyeimbangkan data. Data Diabetes Indian Pima yang telah diseimbangkan kemudian diprediksi dengan metode machine learning seperti metode Bagging, Random Forest, dan XGBoost. Hasil penelitian menunjukkan bahwa performa model machine learning meningkat setelah menangani ketidakseimbangan data dan model terbaik adalah model XGBoost.

Peta Kendali p Berdasarkan Metode Peningkatan Transformasi Akar Kuadrat Rasyid, Riska; Herdiani, Erna Tri; Sunusi, Nurtiti
ESTIMASI: Journal of Statistics and Its Application Vol. 5, No. 1, Januari, 2024 : Estimasi
Publisher : Hasanuddin University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.20956/ejsa.v5i1.18487

When the proportion of nonconformities is small, the effectiveness of the control chart performance becomes inadequate because it has a skewness that causes asymmetryc. Therefore, the Improved Square Root Transformation (ISRT) method is used to construct the attribute control chart to increase the accuracy of the chart control limit which is called the ISRT- control chart. In this study, the effectiveness of the ISRT- control chart perfomance is compared with the control chart after being applied to the data on the number of defects in the newspaper production process at PT. Radar Sulteng Membangun. The results showed that the production process at PT. Radar Sulteng Membangun was not in a statistically controlled and the ARL value obtained on the ISRT- control chart is much smaller than the ARL value for the control chart, so that the ISRT- chart is more effective and sensitive to detecting changes in the production process which produces in a small proportion of nonconformities.

Small Area Estimation for Per Capita Expenditure in Sulawesi Selatan using Empirical Best Linear Unbiased Prediction Miolo, Alya Safira Irtiqa; Sunusi, Nurtiti; Siswanto, Siswanto
InPrime: Indonesian Journal of Pure and Applied Mathematics Vol 6, No 1 (2024)
Publisher : Department of Mathematics, Faculty of Sciences and Technology, UIN Syarif Hidayatullah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15408/inprime.v6i1.33749

Small area estimation (SAE) is an important technique for estimating parameters in regions or sub-populations with limited sample sizes, particularly when direct estimators are inadequate in capturing area-specific information. The Empirical Best Linear Unbiased Prediction (EBLUP) method is one of the SAE parameter estimation approaches, aiming to minimize Mean Square Error (MSE) by incorporating unknown variations of components. In this research, we derive an SAE model parameter estimator and compare its outcomes with both the direct estimator and EBLUP-SAE. The dataset used in this study consists of per capita expenditure data obtained from the March 2019 National Socioeconomic Survey (Susenas) conducted in South Sulawesi, providing a benchmark for assessing household purchasing power. The estimation of SAE parameters was performed using the maximum likelihood method. The results using the EBLUP method reveals that Makassar City recording the highest per capita expenditure at Rp.1,206,352.79 and Jeneponto Regency with the lowest at Rp.1,000,887.29, reflecting significant disparities. Furthermore, the estimated variance of random influence was determined to be 0.010. The study's findings indicate that the EBLUP method outperforms the direct estimation method in estimating per capita expenditure. This is evidenced by the significantly lower MSE value of the EBLUP method, averaging 0.001, compared to the direct estimator’s average MSE value of 0.002. The finding not only emphasizes the reliability of the EBLUP method but also enhances the robustness of socioeconomic analyses and contributes to the advancement of small area estimation techniques. This provides a novelty in understanding regional disparities and informing policy decisions.Keywords: small area estimation, direct estimation, EBLUP, per capita expenditure. AbstrakSmall area estimation (SAE) merupakan metode yang digunakan untuk menduga parameter yang berasal dari area atau sub populasi dengan ukuran sampel yang kecil, ketika estimasi menggunakan penduga langsung tidak mampu menyampaikan informasi area terkait. Metode Empirical Best Linear Unbiased Prediction (EBLUP) merupakan salah satu metode estimasi parameter SAE yang meminimumkan Mean Square Error (MSE) yang dihasilkan dengan asumsi komponen ragam yang tidak diketahui. Penelitian ini bertujuan untuk memperoleh estimator parameter model SAE dan memperoleh perbandingan hasil penduga langsung dan EBLUP-SAE. Data yang digunakan dalam penelitian ini yaitu data pengeluaran per kapita berdasarkan hasil Survei Sosial Ekonomi Nasional (Susenas) Maret 2019 di Sulawesi Selatan, yang berfungsi sebagai tolak ukur untuk menilai kekuatan beli rumah tangga. Estimasi parameter SAE dilakukan menggunakan metode maximum likelihood. Berdasarkan metode EBLUP, diperoleh bahwa nilai pengeluaran per kapita terbesar terjadi di Kota Makassar, yaitu sebesar Rp.1,206,352.79, sedangkan nilai pengeluaran per kapita terkecil terjadi di Kabupaten Jeneponto, yaitu sebesar Rp.1,000,887.29, mencerminkan disparitas yang signifikan. Sementara itu, diperoleh nilai estimasi varians dari pengaruh acak sebesar 0.010. Hasil estimasi dari penelitian ini menunjukkan bahwa metode EBLUP lebih baik dalam melakukan estimasi pengeluaran per kapita dibandingkan metode penduga langsung. Hal ini ditunjukkan dengan nilai MSE dari metode EBLUP menghasilkan rata-rata nilai MSE yang lebih kecil, yaitu sebesar 0.001 dibandingkan dengan rata-rata nilai MSE penduga langsung, yaitu sebesar 0.002. Hal ini tidak hanya menekankan reliabilitas metode EBLUP tetapi juga meningkatkan ketangguhan analisis sosial ekonomi dan berkontribusi pada kemajuan teknik estimasi area kecil. Hal ini memberikan kebaruan dalam pemahaman disparitas regional dan pengambilan keputusan kebijakan.Kata Kunci: small area estimation, penduga langsung, EBLUP, pengeluaran per kapita. 2020MSC: 62J05

Penerapan LASSO Least Trimmed Squares untuk Mengidentifikasi Peubah yang Berpengaruh Penyebaran Penyakit di Sulawesi 138 Selatan Randa, Trigarcia Maleachi; Tinungki, Georgina Maria; Sunusi, Nurtiti
Journal of Mathematics, Computations and Statistics Vol. 6 No. 2 (2023): Volume 06 Nomor 02 (Oktober 2023)
Publisher : Jurusan Matematika FMIPA UNM

Show Abstract | Download Original | Original Source | Check in Google Scholar

Tuberculosis is a chronic infectious disease which is still a public health problem in the world.Indonesia is a country with the third highest tuberculosis burden after India and China. South Sulawesi isone of the provinces that contributes to the high number of tuberculosis cases in Indonesia in 2020. Linearregression analysis can be applied to tuberculosis data to determine the variables that affect the numberof tuberculosis cases in South Sulawesi. Problems that often arise in regression analysis aremulticollinearity problems and outliers in the data. One method that can be used to solve multicollinearityand outlier problems is the LASSO LTS regression. The LASSO LTS regression is a modification of theLASSO regression method based on the LTS estimator of joint regression. The variables in the tuberculosisdata in South Sulawesi have multicollinearity problems and there are outliers, so in this study an approachwith the LASSO LTS method was used to overcome them. The results showed that the LASSO LTS methodcould overcome multicollinearity and outlier problems in estimating regression parameters as evidencedby the highest coefficient of determination of 89.41%.

AVERAGE LINKAGE CLUSTERING METHOD AND MOLECULAR DOCKING STUDY ON DATE PALM (PHOENIX DACTYLIFERA L.) AS POTENTIAL ANTI-ANEMIA AGENT Siswanto, Siswanto; Rasyid, Herlina; Ramadhani, Nur Aliah; Caesar, Nadia Nazwadiah; Sunusi, Nurtiti; Zainuddin, Zaraswati Dwyana
BAREKENG: Jurnal Ilmu Matematika dan Terapan Vol 18 No 4 (2024): BAREKENG: Journal of Mathematics and Its Application
Publisher : PATTIMURA UNIVERSITY

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.30598/barekengvol18iss4pp2459-2470

Anemia, characterized by blood hemoglobin (Hb) levels below the World Health Organization's (WHO) normal limit, remains a significant health concern. Date fruit (Phoenix dactylifera L.) stands out as an herbal plant boasting the highest iron content at 13.7 mg, suggesting its potential as an anti-anemia agent. This study aimed to explore the anti-anemia potential of active compounds in date fruit using average linkage clustering and validated using molecular docking. Compounds from dates were gathered via GC-MS analysis and online databases, totaling 145 compounds—50 from GC-MS and 95 from Knapsack and Dr. Duke databases. Additionally, 5 lead compounds served as positive controls for comparison. SwissADME online servers assessed the compounds' properties, serving as materials for the clustering method. The average linkage clustering method was employed, yielding an excellent dendrogram with a cophenetic correlation of 0.711. Notably, a total of 17 date fruit compounds are in the same cluster as the lead compounds. Molecular docking revealed 4 date palm fruit-derived compounds as potential PHD enzyme inhibitors, promising for anemia treatment. In conclusion, the average linkage clustering method and validation using molecular docking approaches highlight date fruit's potential as an alternative anemia treatment, showcasing the significance of interdisciplinary methodologies in drug discovery.

PERFORMA KLASIFIKASI DATA TIDAK SEIMBANG DENGAN PENDEKATAN MACHINE LEARNING (STUDI KASUS: DIABETES INDIAN PIMA) Aqsha, Masjidil; Sunusi, Nurtiti
Jurnal Matematika UNAND Vol. 12 No. 2 (2023)
Publisher : Departemen Matematika dan Sains Data FMIPA Universitas Andalas Padang

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.25077/jmua.12.2.176-193.2023

Diabetes merupakan suatu penyakit atau gangguan metabolisme kronis dengan multi etiologi yang ditandai dengan tingginya kadar gula darah disertai dengan gangguan metabolisme karbohidrat, lipid, dan protein sebagai akibat insufisiensi fungsi insulin. Faktor risiko diabetes berhubungan dengan status diabetes sesorang. Berbagai pendekatan machine learning menjadi alternatif dalam memprediksi status diabetes. Namun, dalam banyak kasus, data yang tersedia tidak cukup seimbang dalam kelas datanya. Adanya ketidakseimbangan data dapat menyebabkan hasil prediksi menjadi tidak akurat. Tujuan penelitian dalam paper ini adalah untuk mengatasi masalah ketidakseimbangan data dan membandingkan kinerja model dalam memprediksi status diabetes. Secara umum, metode seperti Synthetic Minority Over-sampling Technique (SMOTE) dan Adaptive Synthetic (ADASYN) dapat digunakan untuk menyeimbangkan data. Data Diabetes Indian Pima yang telah diseimbangkan kemudian diprediksi dengan metode machine learning seperti metode Bagging, Random Forest, dan XGBoost. Hasil penelitian menunjukkan bahwa performa model machine learning meningkat setelah menangani ketidakseimbangan data dan model terbaik adalah model XGBoost.Â

Comparison of Negative Binomial Regression Model and Geographically Weighted Poisson Regression on Infant Mortality Rate in South Sulawesi Province Siswanto, Siswanto; Saputra R, Edy; Sunusi, Nurtiti; Ilyas, Nirwan
Indonesian Journal of Statistics and Applications Vol 6 No 2 (2022)
Publisher : Statistics and Data Science Program Study, IPB University, 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.v6i2p170-179

The number of infant mortality cases is an important indicator to assess the quality of a country's public health. A number of studies argue that the case of infant mortality has a close relation to the living area condition and the social status of the parents. Indirectly, the quality of life of babies in a country will impact the nation's quality of life in general. Therefore, many efforts are required to reduce the infant mortality in Indonesia. One of the steps that could be done to overcome this issue is to analyze the causative factors. The statistical method that has been developed for data analysis taking into account current spatial factors is the Geographically Weighted Poisson Regression (GWPR) with a weighted Bisquare kernel function. Based on the partial estimation with the GWPR model, there are seven groups based on significant variables that affect the number of infant deaths in South Sulawesi Province. Of the seven groups formed, the first group is the Selayar Islands where all variables have a significant effect. This needs to be a concern for the South Sulawesi provincial government to improve facilities and infrastructure in the Selayar Islands, of course the location which is very far from the city center can affect access to drug reception, medical personnel and so on. Based on the results of the analysis of the factors that affect the number of infant deaths in South Sulawesi Province using a negative binomial regression approach and GWPR with a bisquare kernel weighting, it can be concluded that the GWPR model used is the best for analyzing the number of infant deaths in South Sulawesi Province because it has an AIC value. The smallest is 167.668.

Small Area Estimation for Per Capita Expenditure in Sulawesi Selatan using Empirical Best Linear Unbiased Prediction Miolo, Alya Safira Irtiqa; Sunusi, Nurtiti; Siswanto, Siswanto
InPrime: Indonesian Journal of Pure and Applied Mathematics Vol. 6 No. 1 (2024)
Publisher : Department of Mathematics, Faculty of Sciences and Technology, UIN Syarif Hidayatullah

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.15408/inprime.v6i1.33749

Small area estimation (SAE) is an important technique for estimating parameters in regions or sub-populations with limited sample sizes, particularly when direct estimators are inadequate in capturing area-specific information. The Empirical Best Linear Unbiased Prediction (EBLUP) method is one of the SAE parameter estimation approaches, aiming to minimize Mean Square Error (MSE) by incorporating unknown variations of components. In this research, we derive an SAE model parameter estimator and compare its outcomes with both the direct estimator and EBLUP-SAE. The dataset used in this study consists of per capita expenditure data obtained from the March 2019 National Socioeconomic Survey (Susenas) conducted in South Sulawesi, providing a benchmark for assessing household purchasing power. The estimation of SAE parameters was performed using the maximum likelihood method. The results using the EBLUP method reveals that Makassar City recording the highest per capita expenditure at Rp.1,206,352.79 and Jeneponto Regency with the lowest at Rp.1,000,887.29, reflecting significant disparities. Furthermore, the estimated variance of random influence was determined to be 0.010. The study's findings indicate that the EBLUP method outperforms the direct estimation method in estimating per capita expenditure. This is evidenced by the significantly lower MSE value of the EBLUP method, averaging 0.001, compared to the direct estimator’s average MSE value of 0.002. The finding not only emphasizes the reliability of the EBLUP method but also enhances the robustness of socioeconomic analyses and contributes to the advancement of small area estimation techniques. This provides a novelty in understanding regional disparities and informing policy decisions.Keywords: small area estimation, direct estimation, EBLUP, per capita expenditure. AbstrakSmall area estimation (SAE) merupakan metode yang digunakan untuk menduga parameter yang berasal dari area atau sub populasi dengan ukuran sampel yang kecil, ketika estimasi menggunakan penduga langsung tidak mampu menyampaikan informasi area terkait. Metode Empirical Best Linear Unbiased Prediction (EBLUP) merupakan salah satu metode estimasi parameter SAE yang meminimumkan Mean Square Error (MSE) yang dihasilkan dengan asumsi komponen ragam yang tidak diketahui. Penelitian ini bertujuan untuk memperoleh estimator parameter model SAE dan memperoleh perbandingan hasil penduga langsung dan EBLUP-SAE. Data yang digunakan dalam penelitian ini yaitu data pengeluaran per kapita berdasarkan hasil Survei Sosial Ekonomi Nasional (Susenas) Maret 2019 di Sulawesi Selatan, yang berfungsi sebagai tolak ukur untuk menilai kekuatan beli rumah tangga. Estimasi parameter SAE dilakukan menggunakan metode maximum likelihood. Berdasarkan metode EBLUP, diperoleh bahwa nilai pengeluaran per kapita terbesar terjadi di Kota Makassar, yaitu sebesar Rp.1,206,352.79, sedangkan nilai pengeluaran per kapita terkecil terjadi di Kabupaten Jeneponto, yaitu sebesar Rp.1,000,887.29, mencerminkan disparitas yang signifikan. Sementara itu, diperoleh nilai estimasi varians dari pengaruh acak sebesar 0.010. Hasil estimasi dari penelitian ini menunjukkan bahwa metode EBLUP lebih baik dalam melakukan estimasi pengeluaran per kapita dibandingkan metode penduga langsung. Hal ini ditunjukkan dengan nilai MSE dari metode EBLUP menghasilkan rata-rata nilai MSE yang lebih kecil, yaitu sebesar 0.001 dibandingkan dengan rata-rata nilai MSE penduga langsung, yaitu sebesar 0.002. Hal ini tidak hanya menekankan reliabilitas metode EBLUP tetapi juga meningkatkan ketangguhan analisis sosial ekonomi dan berkontribusi pada kemajuan teknik estimasi area kecil. Hal ini memberikan kebaruan dalam pemahaman disparitas regional dan pengambilan keputusan kebijakan.Kata Kunci: small area estimation, penduga langsung, EBLUP, pengeluaran per kapita. 2020MSC: 62J05

Title

Found 18 Documents
Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title Search

Found 18 Documents Search

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Abstract

Title

Found 18 Documents
Search