Coronary Heart Disease Detection Using a Combination of Adaptive Synthetic Sampling Approach and Stacking Method on Imbalanced and Incomplete Dataset

EasyChair Preprint 8298

Abstract

Coronary heart disease is one of the most common cardiovascular diseases that lead to death. Therefore, this study proposes an early detection system for coronary heart disease using Framingham dataset with machine learning approach. The system was developed using stacking method of two Machine Learning algorithms, such as Random Forest and Gradient Boosting. It was observed that Framingham dataset has incomplete and imbalanced data classes. Therefore, KNN algorithm and data balancing method were used to solve the problem of incomplete and imbalanced data classes. Two data balancing methods, known as Synthetic Minority Over-sampling Technique (SMOTE) and Adaptive Synthetic Sampling Approach (ADASYN), were compared by evaluating the results of accuracy, precision, recall, and F1-Score. It was discovered that ADASYN with stacking method performed better with accuracy, recall, precision, and f1-score were 90.87%, 89.31%, 92.53%, and 90.89%.

Keyphrases: ADASYN, Coronary Heart Disease, Staking method, machine learning

@booklet{EasyChair:8298,
  author    = {Ahya Radiatul Kamila and Aries Subiantoro},
  title     = {Coronary Heart Disease Detection Using a Combination of Adaptive Synthetic Sampling Approach and Stacking Method on Imbalanced and Incomplete Dataset},
  howpublished = {EasyChair Preprint 8298},
  year      = {EasyChair, 2022}}