This repository contains the R markdown implementation of the final project for the course 'Introduction to Machine Learning in R' by Prof Joseph Yurko as part of academic Credit.

Project Details: The final project for my machine learning course was sponsored by PPG Industries which is a fortune 500 Company, headquartered in Pittsburgh. PPG has given me the opportunity to apply machine learning techniques to study their data on Paint Colors and to see if machine learning techniques can learn patterns associated with top selling paint colors.

Primary goals of the project are associated with learning which INPUTS are important! There are 2 tasks: Regression task: Train models to predict the important paint property, response, as a function of the color model INPUTS. • Want to learn how the color model INPUTS influence the important paint property! • Are the inputs from one color model more influential on predicting the important paint property?

Classification task: Train models to classify if the paint is among the popular paint products sold by PPG based just on color model INPUTS. • Want to learn how the color model INPUTS influence the popularity! • Are the inputs from one color model more influential on the probability the paint is popular?

Project consists of 4 main areas Part i: Exploration • It is always important to explore and study your data before starting any modeling exercise. Part ii: Regression • Fit non-Bayesian and Bayesian linear models. • Train, tune, and assess performance of simple and complex models with resampling. Part iii: Classification • Fit non-Bayesian and Bayesian generalized linear models. • Train, tune, and assess performance of simple and complex models with resampling. Part iv: Interpretation • Identify the best models, most important features, and the hardest to predict paints for the regression and classification tasks

Unfortunately, I cannot share the data, nor the predictions.