Data Mining for Lifestyle Risk Factors Associated with Overweight and Obesity among Adolescents

Abstract

Data mining techniques have been applied to many areas in the business world and our daily life, including healthcare and clinical health services. One of the mostly watched health problems is obesity and overweight, particularly for children and adolescents. In this paper, we try to find the most significant lifestyle risk factors associated with overweight and obesity among high school students in the US. Lifestyle survey data from the 2011 National Youth Risk Behavior Survey (YRBS) was used with the students' body weight statuses, overweight or obesity, considered as two target variables. Both logistic regression models and decision tree models were created for each target variable. Both the logistic regression and decision tree method show that frequently doing physical activity and having breakfast everyday were protective factors against being overweight or obese. Smoking and drinking sugar-sweeten beverage frequently were found to be associated with an increased risk to be obese.