| dc.description.abstract | In recent times, healthcare has become one of the most significant global concerns,
especially following the COVID-19 pandemic. Chronic diseases, such as cardiovascular
diseases (CVDs), remain a leading cause of mortality worldwide, particularly in the United
States. CVDs are usually appropriately recognized based on indicators measured in the
hospital, but due to people's busy schedules, many are unwilling to spend time and effort for
check-ups in the hospital. Lifestyle habits are assumed to play a role in the development of
CVDs, making it crucial to modify these behaviors to predict and prevent these diseases.
Previous studies have used machine learning to predict disease risk based on various
factors. However, these studies may not fully capture the dynamic relationship between
lifestyle data and the presence of cardiovascular diseases. To address this, this thesis aims to
build on previous findings and improve the accuracy of predicting CVDs by analyzing data
from the NHANES dataset, which combines information on lifestyle habits and health
indicators, using machine learning techniques such as CatBoost, Gradient Boosting,
XGBoost, Linear Regression, and Support Vector Machines. The ultimate objective is to
identify the relationship between lifestyle habits and CVDs, and thus reduce the number of
potential patients. | en_US |
