Daniel-DS-dev/Expresso-Churn-Challenge-Data-Science-Nigeria-

# Expresso-Churn-Challenge-Data-Science-Nigeria ### This repository contains my solution to the Pre-Bootcamp hackathon organised by Data Science Nigeria (DSN) on Zindi, from 8 August — 22 August, 2020. (link to hackathon: zindi.africa) ### Main Objective: Develop a predictive model that determines the likelihood for a customer to churn, i.e. stop purchasing airtime and data from Expresso. ### Evaluation Metric: Log loss ### Machine Learning Models/Algorithms Used: Random Forest, XGBoost, Logistic Regression, Gradient Boost, LGBM and CatBoost Algorithm. Catboost algorithm gave me the best result. ## This solution gave me a log loss score of: 0.247057826917659. * Expresso is an African telecommunications company that provides customers with airtime and mobile data bundles. The objective of this challenge is to develop a machine learning model to predict the likelihood of each Expresso customer “churning,” i.e. becoming inactive and not making any transactions for 90 days. * This solution will help Expresso to better serve their customers by understanding which customers are at risk of leaving * About Data Science Nigeria (www.datasciencenigeria.org): # An extensive explanation about treating missing data is included in the notebook.