An Imputation for Missing Data Features Based on Fuzzy Swarm Approach in Heart Disease Classification

Computational intelligence methods have been broadly applied to define the important features for Heart Disease classification. Nonetheless, imprecise features data such as no values and missing values can affect quality of classification results. Nevertheless, the other complete features are still capable to give information in certain features. Therefore, an imputation approach based on Fuzzy Swarm is developed in preprocessing stage. It will help to fill in the missing values by cluster the complete candidates and optimizes it using Particle Swarm Optimization (PSO). Then, the complete dataset is trained in classification algorithm, Decision Tree. The experiment is trained with Heart Disease dataset and the performance is analyzed using accuracy, precision, and ROC values. Results show that the performance of Decision Tree is increased after the application of Fuzzy Swarm for imputation.