Volume 18, No. 6, 2021
An Efficient Pre-Processing Method Using Optimization Techniques For Heart Disease Prediction
V. Chezhiyan , Dr. D.J. Evanjaline
Abstract
The recent technology developments and innovations improves the life style of people through smart applications, sensors, wireless communication networks, etc., for all those technologies internet is the backbone and the information processing like accessing, distributing the necessary information is achieved through Internet of Things (IoT). IoT supports multidisciplinary applications as an active entity in engineering, science and business discipline. Based on the user preference these applications and its services could be framed in IoT. Human daily activity recognition using mobile personal sensing technology plays a central role in the field of pervasive healthcare. Handling of huge volume of sensor data is a crucial issue in this domain for an effective decision-making system. In this research work, an effective preprocessing method is proposed using wrapper-based feature selection techniques. Harris Hawks Optimization (HHO) and Genetic Algorithm (GA) are hybridized to get the most predominant features for the classification of heart disease where the data is obtained by the IoT wearable devices. The performance of the proposed pre-processing method is analysed with the existing feature selection techniques using different classifiers like Random Forest (RF), Gradient Boosting Tree (GBT) and Support Vector Machine (SVM) with various evaluation metrics like Accuracy, Precision, Recall and error rates.
Pages: 6574-6588
Keywords: Healthcare, Internet of Things (IoT), Optimization algorithms, Feature Selection, Classification.