• Used PySpark to perform exploratory data analysis in order to estimate churn using a 12GB dataset stored in an S3 bucket • Identified crucial features using feature engineering to train ML models on the cloud with the highest test accuracy of 84%
• Used PySpark to perform exploratory data analysis in order to estimate churn using a 12GB dataset stored in an S3 bucket • Identified crucial features using feature engineering to train ML models on the cloud with the highest test accuracy of 84%