A capstone project that helps a school develop a tool to assist in identifying key contributors to obesity. The project will employ techniques to identify factors contributing to obesity based upon publicly available data. Attached is the capstone template, topic form, and the rubric on what requirements must be met.
Descriptive
method: Visualizations and data summary will be used to get a high-level
overview of the variable and their distributions. Also, correlation analysis to
measure the strength and direction of relationships between the different
variables and obesity.
Predictive/Prescriptive
method: A machine learning model will be developed to build predictive
regression models to predict obesity rates based on identified key factors.
IMPLEMENTATION and
EVALUATION:
a.
Feature
Selection and Engineering – identify the independent variables that are likely
to influence obesity based on knowledge and exploratory dat analysis
b.
Model
selection – implement regression predicitive modeling
c.
Model
training – train the models on the dataset using and optimize hyperparameters
to achieve a better model performance
d.
Once
the code is complete and provides interpretation of results, supporting
documentation will be written and a working version of the product will be
given to WGU Health and Wellness
e.
Continuous
Improvement – As the client WGU Health and Wellness test the product, update
and retrain the predictive models to keep the insights relevant and up-to-date