Heart Disease Prediction using Machine Learning Algorithms
1Ch.Varshith, 2H. Varshith, 3D. Varshitha, 4M. Varshitha, 5T. Varshitha, 6G. Varun, 7Dr. K. Satish
123456Student, 7 Professor
School of Engineering – IIIrd year AI&ML-ZETA
12111cs020620@mallareddyuniversity.ac.in 2 2111cs020621@mallareddyuniversity.ac.in
32111cs020622@mallareddyuniversity.ac.in 4 2111cs020623@mallareddyuniversity.ac.in
52111cs020624@mallareddyuniversity.ac.in 6 2111cs020625@mallareddyuniversity.ac.in
Department of Artificial Intelligence & Machine Learning
Malla Reddy University, Kompally, Hyderabad, India
Abstract
Presently, health conditions are on the rise primarily due to lifestyle factors and genetic predispositions. Among these, heart disease has notably surged, posing a significant risk to people's lives. Each person possesses unique benchmarks for vital health indicators like blood pressure, cholesterol levels, and pulse rates. However, according to established medical standards, normal values typically fall within certain ranges: Blood pressure ideally registers at 120/90, cholesterol between 100-129 mg/dL, pulse rate around 72, fasting blood sugar level at 100 mg/dL, heart rate ranging from 60-100 beats per minute (bpm), normal ECG readings, and major vessel widths spanning from 25 mm (1 inch) in the aorta to a mere 8 μm in capillaries.
This study explores various classification techniques employed to predict the risk levels of individuals based on parameters such as age, gender, blood pressure, cholesterol levels, and pulse rate. The "Disease Prediction" system relies on predictive modeling to anticipate a user's disease risk by analyzing symptoms provided as input. By evaluating user-input symptoms, the system computes the probability of specific diseases as an output. Disease prediction involves the implementation of five techniques: Naïve Bayes, KNN, Decision Tree,
Linear Regression, and Random Forest Algorithms. These methods gauge the likelihood of an individual developing a particular ailment. Consequently, the collective average prediction accuracy probability stands at an impressive 83%.
Keywords: Naïve Bayes, KNN, Decision Tree, Linear Regression & Random Forest.
