<![CDATA[According to WHO (World Health Organization) data, heart disease accounts for one-third of all fatalities worldwide. Heart failure causes the death of approximately 17.9 million people worldwide and is more prevalent in Asia. With the use of technology, biostatistical analysis data may be processed using data mining techniques to uncover correlation patterns for each data from previous data that can then be used to forecast diseases based on these patterns. Algorithms with high accuracy can at least help medical experts prevent heart failure deaths. The ultimate goal of this research is to compare the algorithms and methods of Standard Scalers and robust scalers on datasets that have implemented the SMOTE method and without SMOTE to find algorithms and techniques with the highest accuracy and performance from the best algorithms using ROC curve values and AUC values. In this study, researchers will use Logistic Regression, Random Forest, and Perceptron. Comparing the Standard Scaler and Robust Scaler methods as data standardization. And using the SMOTE technique and not using SMOTE to overcome unbalanced classes. As a result, the Random Forest algorithm with SMOTE technique and Standard Scaler is suitable for classifying and predicting heart failure.]]>
Copyrights © 2022
