PENERAPAN METODE RESAMPLING DALAM MENGATASI IMBALANCED DATA PADA DETERMINAN KASUS DIARE PADA BALITA DI INDONESIA (ANALISIS DATA SDKI 2017)

Authors

  • Andriansyah Muqiit WS Politeknik Statistika STIS
  • Rani Nooraeni Politeknik Statistika STIS

DOI:

https://doi.org/10.24252/msa.v8i1.13452

Abstract

Data yang memiliki rasio yang tidak berimbang antara data satu dengan data lainnya dapat dikatakan sebagai imbalanced. Metode Synthetic Minority Oversampling Technique (SMOTE) merupakan metode yang populer diterapkan dalam rangka menangani imbalanced. Pada penelitian ini penulis ingin membandingkan model data sebelum dan data sesudah dilakukan oversampling menggunakan analisis regresi logistik berganda. Data yang digunakan adalah data sekunder dari hasil Survei Demografi dan Kesehatan (SDKI) tahun 2017. Variabel respon (dependent) yang digunakan adalah balita mengalami diare dalam dua minggu terakhir sebelum pencacahan. Variabel penjelas dikelompokkan menjadi dua yaitu faktor sosio-demografi dan faktor lingkungan. Variabel penjelas yang termasuk ke dalam faktor sosio-demografi antara lain jenis kelamin balita, umur balita, umur ibu, dan tingkat pendidikan ibu. Sedangkan variabel penjelas yang merupakan faktor lingkungan antara lain sumber air minum, jenis fasilitas toilet, jenis lantai rumah utama, dan daerah tempat tinggal. Berdasarkan penelitian yang telah dilakukan dapat disimpulkan bahwa model dengan data hasil SMOTE pada uji parsial menghasilkan lebih banyak variabel yang signifikan dibandingkan dengan model dengan data tanpa SMOTE.

References

[1] Barro, Rossi Azmatul, dkk. 2013. Penerapan Synthetic Minority Oversampling Technique (SMOTE) terhadap Data Tidak Seimbang pada Pembuatan Model Komposisi Jamu. Jurnal Xplore Vol.1, 2013, 1-6.

[2] Blagus, Rok dan Lara Lusa. 2013. SMOTE for High-Dimensional Class-Imbalanced Data. BMC Bioinformatics 2013, 14:106.

[3] Chawla, Nitesh V., dkk. 2002. SMOTE: Synthetic Minority Over-sampling Technique. Journal of Artificial Intelligence Research 16, 2002, 321-357.

[4] Chawla, Nitesh V., dkk. 2018. SMOTE for Learning from Imbalanced Data: Progress and Challenges, Marking the 15-year Anniversary. Journal of Artificial Intelligence Research 61, 2018, 863-905.

[5] Ganganwar, Vaishali. 2012. An Overview of Classification Algorithms for Imbalanced Datasets. International Journal of Emerging Technology and Advanced Engineering Vol.2 Issue 4, April 2012, 42-47.

[6] Rahayu, Sri, dkk. 2017. Analisis Perbandingan Metode Over-Sampling Adaptive Synthetic-Nominal (ADASYN-N) dan Adaptive Synthetic-kNN (ADASYN-kNN) untuk Data dengan Fitur Nominal-Multi Categories. Jurnal CITEE, 2017.

[7] Siringoringo, Rimbun. 2018. Klasifikasi Data Tidak Seimbang Menggunakan Algoritma SMOTE Dan K-Nearest Neighbor. Jurnal ISD Vol.3 No.1 Januari - Juni 2018.

[8] Skryjomski, Przemyslaw dan Bartosz Krawczyk. 2017. Infuence of Minority Class Instance Types on SMOTE Imbalanced Data Oversampling. Proceedings od Machine Learning Research Vol.74, 2017, 7-21.

[9] Untoro, Meida Cahyo dan Joko Lianto Buliali. 2018. Penanganan imbalance class data laboratorium kesehatan dengan Majority Weighted Minority Oversampling Technique. Jurnal Ilmiah Teknologi Sistem Informasi 4 (1), 2018, 23-29.

[10] Zheng, Zhuoyuan, dkk. 2015. Oversampling Method for Imbalanced Classification. Computing an Informatics Vol.34 No. 5 2015, 1017-1037.

Downloads

Published

2020-07-06

How to Cite

[1]
A. Muqiit WS and R. Nooraeni, “PENERAPAN METODE RESAMPLING DALAM MENGATASI IMBALANCED DATA PADA DETERMINAN KASUS DIARE PADA BALITA DI INDONESIA (ANALISIS DATA SDKI 2017)”, MSA, vol. 8, no. 1, pp. 19–27, Jul. 2020.

Issue

Vol. 8 No. 1 (2020): Volume 8 Nomor 1

Section

Artikel

License

Jurnal MSA applies the Creative Commons Attribution 4.0 International License (CC BY 4.0) to its published articles.

Under this license, readers are free to read, download, copy, distribute, print, search, link to the full texts of articles, and reuse the published materials for any lawful purpose, including commercial purposes, provided that appropriate credit is given to the original author(s) and the source of publication.

The terms of this license are as follows:

  1. Attribution: Users must give appropriate credit to the author(s), cite the original article, acknowledge Jurnal MSA as the source of publication, provide a link to the license, and indicate whether any changes have been made.

  2. Sharing and Redistribution: Users may copy and redistribute the published material in any medium or format, provided that the license conditions are respected.

  3. Adaptation and Reuse: Users may remix, transform, translate, and build upon the published material for any purpose, including commercial purposes, subject to the attribution requirements of CC BY 4.0.

  4. Copyright Ownership: Authors retain the copyright of their published works. By publishing in Jurnal MSA, authors grant permission for their articles to be distributed and reused under the terms of the CC BY 4.0 license.

  5. No Additional Restrictions: Users may not impose additional legal or technological restrictions that prevent others from exercising the rights granted by this license.

  6. Third-Party Materials: Materials incorporated into an article that are owned by third parties may be subject to separate copyright or licensing terms. Users are responsible for obtaining any additional permissions required for their reuse.

The CC BY 4.0 license does not remove the obligation to acknowledge the original authors and source. Users must comply with the applicable attribution requirements when reusing or redistributing published articles.

For the complete license terms, please visit:

Creative Commons Attribution 4.0 International (CC BY 4.0)
https://creativecommons.org/licenses/by/4.0/

By accessing and reusing articles published in Jurnal MSA, users agree to respect the applicable license conditions.

Most read articles by the same author(s)