Using numerous linear and non-linear algorithms to determine top model for predicting house prices in Boston.

This project forecasts sales for specific electronic products. Random Forest, Support Vector Machines RBF Kernel, and Gradient Boosting were algorithms of choice for this regression problem.

It's all about marketing, inventory, and maximizing profits.

Forecasting energy use for a client's residential home using time series linear regression and Holt Winters modeling. Source data contains over 2 million minute observations of energy in Watts over 5 year time span.

This market basket analysis not only finds product associations, but also provides business insights based on product transaction relationships. This comes in really handy for recommender systems, cross-selling, and email promotions.

The objectives are to conduct an in-depth time series analysis of energy records from 2007 to 2010 for a law firm whose client claims to have not been occupying the residence during an undisclosed event. Time Series processing Tidyverse data wrangling ggplot2 visualizing