Exploratory Analysis of Heart Attack and Breast Cancer Early Stage Prediction

This paper involves the development of a web based platform with an integrated machine learning algorithm to address problems pertaining to heart disease and breast cancer. It enables users to select either the heart attack or breast cancer datasets and use machine learning models such as Logistic Regression, K-Nearest Neighbor (KNN), Support Vector Machine (SVM), Random Forest and XG Boost. The website provides basic information about the datasets such as the total number of records in a dataset, number of classes in a dataset, and also allows users to generate graphs for better visualization of the data. The evaluation of the performance of each of the models is based on various metrics such as accuracy, F1 score, precision, recall and mean squared error. This way it is possible to assess the performance of the models in relation to each of the available data sets which makes it easy for users to pick the most effective model. There are also inputs for the users which would be their health indicators like age, sex, level of cholesterol, heart rate among others. Using these values as well as the picked model, the website estimates the possibility of the user developing heart disease or breast cancer. Apart from predicting health risks, there is an emphasis on the evaluation of health risks by different machine learning models and thus assists users in finding the most accurate prediction model regarding their healthcare problems. Overall, the website provides both predictive healthcare insights and a deeper understanding of how various algorithms perform in real-world scenarios.