Diabetes Mellitus Disease Prediction Using Machine Learning Classifiers with Oversampling and Feature Augmentation

Abstract

The technical improvements in healthcare sector today have given rise to many new inventions in the field of artificial intelligence. Patterns for disease identification are carried out, and the onset of prediction of many diseases is detected. Diseases include diabetes mellitus disease, fatal heart diseases, and symptomatic cancer. There are many algorithms that have played a critical role in the prediction of diseases. This paper proposes an ML based approach for diabetes mellitus disease prediction. For diabetes prediction, many ML algorithms are compared and used in the proposed work, and finally the three ML classifiers providing the highest accuracy are determined: RF, GBM, and LGBM. The accuracy of prediction is obtained using two types of datasets. They are Pima Indians dataset and a curated dataset. The ML classifiers LGBM, GB, and RF are used to build a predictive model, and the accuracy of each classifier is noted and compared. In addition to the generalized prediction mechanism, the data augmentation technique is also used, and the final accuracy of prediction is obtained for the classifiers LGBM, GB, and RF. A comparative study and demonstration between augmentation and non-augmentation are also discussed for the two datasets used in order to further improve the performance accuracy for predicting diabetes disease.

A Deep Learning-Based Diabetes Diagnosis Model on PIMA Image Dataset

“…The suggested model yields typical outcomes in the identification of diabetes mellitus. Before applying machine learning techniques, it was advised to utilize data augmentation on the PIMA dataset to prevent under-sampling because of the tiny dataset [18]. When combined with augmented datasets, machine learning algorithms significantly improve their ability to forecast diabetes mellitus.…”

Chronic Diseases Prediction Using Machine Learning With Data Preprocessing Handling: A Critical Review

“…Based on research on the prediction of chronic disease, here are some names of methods used for handling imbalance data. (1) oversampling techniques: SMOTE [70], [85]- [88]; ADASYN [60], [89], [90]; ROS [5], [20]; orchard SMOTE [91]; SMOTE-Tomek [92]; TimeGAN49 [93]; and SVM-SMOTE [94]. (2) undersampling techniques such as: Tversky similarity [95], near miss [60], RUS [60], and NCL [89].…”

“…More tolerant to overfitting compared to a single decision tree XGBoost [63] max_depth: 1-10, gamma: [0, 0.4-1], min_child_weight: [1][2][3][4][5][6]8,10] Boosting for accuracy prediction C4.5 [206] Criterion: gini, max_depth: none, n_estimators: 150 Automatically identifying key risk factors associated with stroke CatBoost [10] learning_rate: 0.03, class_weight: 1, iterations: 100, depth: 6…”

“…Machine learning can be used for data mining in the healthcare sector [4]. Applying machine learning in health data can help predict if a patient might have six chronic diseases: diabetes mellitus [5], [6]; cancer [7], [8]; stroke [9], [10]; hypertension [11], [12]; kidney failure [13], [14]; and heart issues [15], [16].…”

Diabetes Mellitus Disease Prediction and Type Classification Involving Predictive Modeling Using Machine Learning Techniques and Classifiers

“…It is calculated using 2 , where y1 is the actually calculated value and y1 ′ is the fnally value that is predicted by the model. So y1 ′ is replaced with G n (X) which represents the target value [24]. It is mathematically given as follows:…”