Original Article
Split Viewer
International Journal of Fuzzy Logic and Intelligent Systems 2023; 23(2): 140-151
Published online June 25, 2023
https://doi.org/10.5391/IJFIS.2023.23.2.140
© The Korean Institute of Intelligent Systems
Imbalanced Learning in Heart Disease Categorization: Improving Minority Class Prediction Accuracy Using the SMOTE Algorithm
Mediana Aryuni1 
, Suko Adiarto2 , Eka Miranda1 , Evaristus Didik Madyatmadja1 , Albert Verasius Dian Sano3 , and Elvin Sestomi1
1Department of Information Systems, Bina Nusantara University, Jakarta, Indonesia
2Department of Cardiology and Vascular Medicine, University of Indonesia, Jakarta, Indonesia
3Department of Computer Science, Bina Nusantara University, Jakarta, Indonesia
Correspondence to :
Mediana Aryuni (mediana.aryuni@binus.ac.id)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
In the field of medical data mining, imbalanced data categorization occurs frequently, which typically leads to classifiers with low predictive accuracy for the minority class. This study aims to construct a classifier model for imbalanced data using the SMOTE oversampling algorithm and a heart disease dataset obtained from Harapan Kita Hospital. The categorization model utilized logistic regression, decision tree, random forest, bagging logistic regression, and bagging decision tree. SMOTE improved the model prediction accuracy with imbalanced data, particularly for minority classes.
Keywords: Heart disease, Prediction, Imbalanced, Accuracy, SMOTE
Conflict of Interest
No potential conflicts of interest relevant to this article were reported.
Biographies
PDF
Standard view
Export citation
Share Article
Original Article
International Journal of Fuzzy Logic and Intelligent Systems 2023; 23(2): 140-151
Published online June 25, 2023 https://doi.org/10.5391/IJFIS.2023.23.2.140
Copyright © The Korean Institute of Intelligent Systems.
Imbalanced Learning in Heart Disease Categorization: Improving Minority Class Prediction Accuracy Using the SMOTE Algorithm
Mediana Aryuni1 , Suko Adiarto2 , Eka Miranda1 , Evaristus Didik Madyatmadja1 , Albert Verasius Dian Sano3 , and Elvin Sestomi1
1Department of Information Systems, Bina Nusantara University, Jakarta, Indonesia
2Department of Cardiology and Vascular Medicine, University of Indonesia, Jakarta, Indonesia
3Department of Computer Science, Bina Nusantara University, Jakarta, Indonesia
Correspondence to:Mediana Aryuni (mediana.aryuni@binus.ac.id)
This is an Open Access article distributed under the terms of the Creative Commons Attribution Non-Commercial License (http://creativecommons.org/licenses/by-nc/3.0/) which permits unrestricted noncommercial use, distribution, and reproduction in any medium, provided the original work is properly cited.
Abstract
In the field of medical data mining, imbalanced data categorization occurs frequently, which typically leads to classifiers with low predictive accuracy for the minority class. This study aims to construct a classifier model for imbalanced data using the SMOTE oversampling algorithm and a heart disease dataset obtained from Harapan Kita Hospital. The categorization model utilized logistic regression, decision tree, random forest, bagging logistic regression, and bagging decision tree. SMOTE improved the model prediction accuracy with imbalanced data, particularly for minority classes.
Keywords: Heart disease, Prediction, Imbalanced, Accuracy, SMOTE
Fig 1.
Download
Data preprocessing.
Fig 2.
Download
Model development.
Fig 3.
Download
ROC AOC curve example [
Fig 4.
Download
ROC curves of five classifier models without SMOTE algorithm: (a) LR, (b) DT, (c) RF, (d) bagging LR, (e) bagging DT.
Fig 5.
Download
ROC curves of five classifier models with SMOTE algorithm: (a) LR, (b) DT, (c) RF, (d) bagging LR, (e) bagging DT.
-
Table 1 . Attribute description.
Attribute name Description gender male or female age age in years eritrosit the number of red blood cells in the patient’s blood vessels hematokrit percentage of red blood cells in blood hemoglobin the amount of oxygen and iron-carrying protein in the patient’s body hermch the average mass of hemoglobin per red blood cell in a blood sample from a patient khermchc the patient’s average hemoglobin concentration per red blood cell leukosit white blood cell number in the patient’s blood vessels trombosit the number of platelets in the patient’s blood for blood clotting diagnose class label (class target)
-
Table 2 . Performance of each model without SMOTE algorithm.
Model Performance measurement Precision (%) Recall (%) F1-score (%) AUC (%) LR 47 50 49 64 DT 70 60 63 62 RF 71 58 61 70 Bagging LR 72 51 50 69 Bagging DT 98 66 73 97
-
Table 3 . Performance of each model with SMOTE algorithm.
Model Performance measurement Precision (%) Recall (%) F1-score (%) AUC (%) LR 66 66 66 72 DT 76 75 75 82 RF 78 78 78 87 Bagging LR 66 66 66 73 Bagging DT 83 83 83 92
-
Table 4 . Performance of each model before and after using SMOTE algorithm in predicting minority class.
Model Performance measurement using SMOTE Precision (%) Recall (%) F1-score (%) Before After Before After Before After LR 9 66 67 67 15 66 DT 10 70 81 84 18 76 RF 14 75 82 81 24 78 Bagging LR 9 66 67 67 15 66 Bagging DT 19 81 76 76 30 78
Article Tools
Metrics
Share this article on :
Related articles in IJFIS
-
Gaussian-Based SMOTE Algorithm for Solving Skewed Class Distributions
Hansoo Lee, Jonggeun Kim, and Sungshin Kim
Int. J. Fuzzy Log. Intell. Syst. 2017; 17(4): 229-234
https://doi.org/10.5391/IJFIS.2017.17.4.229
Gaussian-Based SMOTE Algorithm for Solving Skewed Class Distributions
Hansoo Lee, Jonggeun Kim, and Sungshin Kim
Int. J. Fuzzy Log. Intell. Syst. 2017; 17(4): 229-234 https://doi.org/10.5391/IJFIS.2017.17.4.229
