Description

Coronary heart condition (CHD) is one among the main causes of incapacity in adults yet united of the most causes of death within the developed countries. though vital progress has been created within the designation and treatment of CHD, more investigation is still required. the target of this study was to develop a data-mining system for the assessment of heart event-related risk factors targeting within the reduction of CHD events. the chance factors investigated were: 1) before the event: a) nonmodifiable—age, sex, and case history for premature CHD, b) modifiable—smoking before the event, history of cardiovascular disease, and history of diabetes; and 2) when the event: modifiable—smoking when the event, systolic blood pressure, pulse pressure level, total sterol, highdensity lipoprotein, LDL, triglycerides, and aldohexose. The events investigated were: myocardial infarct (MI), percutaneous coronary intervention (PCI), and arteria coronaria bypass graft surgery (CABG). a complete of 528 cases were collected from the Paphos district in Cyprus, most of them with over one event. Data-mining analysis was allotted mistreatment the C4.5 decision tree formula for the aforesaid 3 events mistreatment five completely different cacophonic criteria. the foremost necessary risk factors, as extracted from the classification rules analysis were: 1) for MI, age, smoking, and history of cardiovascular disease; 2) for PCI, case history, history of hypertension, and history of diabetes; and 3) for CABG, age, history of cardiovascular disease, and smoking. Most of those risk factors were conjointly extracted by different investigators. the best percentages of correct classifications achieved were sixty six, 75%, and 75% for the MI, PCI, and coronary bypass surgery models, severally. it’s anticipated that data processing may facilitate within the identification of high and low risk subgroups of subjects, a determinant for the choice of therapy, i.e., medical or surgical. Also we implement the Pruning algorithm in order to reduce the size of the Decision Tree for more understanding. This will reduce the size of the decision tree by removing the attribute with minimum information and construct the tree with attributes which give higher information for analysis. In our System we use Basic Pruning algorithm which prunes the attribute has information gain below the threshold level and construct the decision tree with attributes which gives information greater than the threshold level. However, more investigation with larger datasets remains required.