Classification Machine Learning Project of a cardiovascular diseases dataset.
Web application of the Project: https://cardiovascular-datascience-robsonserafim.streamlit.app/
I strongly believe that the primary objective of data science is to contribute to society by providing valuable insights. Therefore, the application of data science in the healthcare industry holds immense significance and offers numerous benefits to society. With this in mind, as I delve deeper into the field of data science, I have chosen to explore healthcare data as the foundation for my data science project. My project aims to encompass various components, including Exploratory Data Analysis, developing a classification Machine Learning Model, and deploying the model through a Streamlit web application. By undertaking this project, I aim to harness the power of data science to generate meaningful outcomes in the healthcare sector.
Cardio Vascular Disease its a relevant data source that relates risk factors and cardiovascular disease. Facing the fact that, according to the World Health Organization (WHO), cardiovascular diseases are the leading cause of death in the world, causing the death of about 17.7 million people every year, I chose this dataset to develop an exploratory analysis. The dataset used is called Cardiovascular Disease dataset and the project created was based on implementing an exploratory analysis aiming to demonstrate and provide insights about different variables related to the presence of heart diseases in the analyzed patients. To achieve this goal, I used the programming language Python, libraries for data analysis such as pandas and numpy and for graphical visualization I used mainly the libraries matplotlib and seaborn.
Furthermore, the main goal of the project was to use a Classification Machine Learning Model to predict whether the patient has (flag 1) or not (flag 0) cardiovascular disease, based on all his health parameters present in the dataset. After performing a comparative ranking between different classification models, the model chosen to perform the deployment was XGBoost Classifier.**
Robson de Castro Serafim