Machine Learning

Jupyter Notebook Advertising Spend Regression Models A comparison of different models in the prediction of sales based on channel advertising channels spend. Models include Multiple Regression, Random Forest, Gradient Boosting, Neural Network and more

Assumptions Sub-plotting: Sub-plotting assumptions for linearity, normality, and homoscedasticity.

Classification Models: Data Process Mapping out the steps of the process from planning, and analyzing, to model selection, building, and measuring.

Building a Human Resources Employee Attrition Model Building a human resources model to predict employee prediction. The model uses the Random Forests algorithm and compares its results with the Logistical Regression algorithm.

Github: Jupyter Notebooks Titanic Binary Classification Models Full Exploratory Data Analysis and Classification Model building for prediction of Titanic survivor data (Kaggle Task)

Building a Predictive Customer Lifetime Model Walk through trying to build a predictive customer lifetime model with Logistical Regression and the Contoso Retail data warehouse. It doesn’t go well, so there’s a challenge ahead!

K-Means Clustering – Supermarket Sales by Customer An example of using clustering to identify potential customer segments