![]() | Only 14 pages are availabe for public view |
Abstract Heart disease is a serious disease that leads to death; there are different types ad categories of heart disease, such as coronary, cardiovascular and cardiomyopathic. Smoking, hypertension and obesity rank among the most important risk factors that influence heart performance; Heart disease is associated with functional problems of the heart such as irregular heart rhythms, which increase the risk of heart attack occurrence and other heart problems. There is a significant difference between the symptoms of heart disease and the factors leading to it because the symptoms of the disease consist of a set of signs that the person exhibits when affected by heart disease. Most of the previous researches used symptoms to predict heart diseases that determines the type and degree of a person’s heart disease. The factors that lead to the occurrence of heart disease (risk factors) represent a group of diseases that affect the person or some parts of their behavior, which in turn lead to a person suffering from heart disease; these risk factors can be used as attributes to build a system that can predict heart disease. Therefore, the main objective of this work was to build an Adaptive Heart Disease Behavior- Based Prediction System (AHDBP) using different classification algorithms, to identify and correct all the symptoms of heart disease used in previous studies in this field, and to validate the results using the suitable measuring standards. In this research, a set of new risk factors attributes based on World Health Organization (WHO) reports for 2018 are used to build the prediction system. The data set is classified by three basic classification techniques: Decision Tree, Naive Bayes and Neural Networks. The accuracy of the system is tested by different evaluation techniques. The accuracy of the classification techniques was as follows: Decision Tree 90.:14%, aive Bayes 91.54%, and Neural Networks 94.91%. Neural networks can predict heart disease better than other techniques. The Chi square method has also been applied to determine the difference between the expected and the observed results, and the proposed system proved its accuracy at 86.54%. TIle proposed prediction framework is designed to help doctors in heart diseases prediction, where the accuracy of heart diseases will be improved by using neural network. |