The aim of the analysis is to understand behaviours that are most predictive of a new user starting and staying active on Product-X. The Growth team at Company-X is interested in month one retention, defined as whether a user remains active after signup. We will use the datasets to understand what factors are the best predictors of retention, and offer suggestions to operationalize these insights and help Company-X! This report highlights different methodologies used in Binary Classification, often used in retention analysis, and in this case to specifically predict and analyze factors that possibly affect month one user retention. I have used Logistic Regression, Decision Trees, and Random Forests to build predictive models and Precision, Recall and the F1 Statistic as metrics for them.
This serves as a good introduction to Churn Prediction, Binary Classification, Big data cleaning and processing. You can find the code and detailed analysis here