To determine whether behavioral factors differ among metabolic conditions and self-reported health, and to determine whether self-reported health is a valid predictor of metabolic syndrome (MetS). A total of 2997 individuals (≥ 40 years old) were selected from four biennial U.S. National Health and Nutrition Examination Surveys (2007–2014). A set of weighted logistic regression models were used to estimate the odds ratios (ORs) and 95% confidence intervals (CIs)Individuals with light physical activity are more likely to have MetS and report poor health than those with vigorous physical activity with OR = 3.22 (95% CI: 2.23, 4.66) and 4.52 (95% CI: 2.78, 7.33), respectively. Individuals eating poor diet have greater odds of developing MetS and reporting poor health with OR = 1.32 (95% CI: 1.05, 1.66) and 3.13 (95% CI: 2.46, 3.98). The aforementioned relationships remained significant after adjustment for demographic and socio-economic status. A potential intervention strategy will be needed to encourage individuals to aggressively improve their lifestyle to reduce MetS and improve quality of life. Despite the significant association between self-reported health with MetS, a low sensitivity indicated that better screening tools for MetS, diabetes and cardiovascular disease are essential.

This project is dedicated to the problem of vehicle classification using inductive loop sensors. We created the dataset that contains more than 11000 labeled inductive loop signatures collected at different times and from different parts of the world. Multiple classification methods and their optimizations were employed to the vehicle classification. Final model that combines K-nearest neighbors and logistic regression achieves 94\% accuracy on classification scheme with 9 classes. The vehicle classifier was implemented in C++.