Building a churn predictor using Hadoop
In this section, we will apply the knowledge gained in the previous section to build a churn predictor that can scale up for use with very large datasets. Hadoop is particularly suited to the processing of very large datasets.
Synthetic data generation tools
To build our churn prediction using Hadoop, we will work with a synthetic dataset. Synthetic datasets are good for testing and learning purposes because they do not carry the risk of personal information leakage, and they are simple to obtain or create. But they also carry the limitation that, unlike a real-life dataset, they never contain the truth. Synthetic data offers a good alternative to real life datasets if you want to get started quickly.
Some ...
Get Hadoop Blueprints now with the O’Reilly learning platform.
O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.
