This pipeline calculates the LTV (life-time value) value of a mobile application user by analyzing the in-app events within the first 3 days after registration. LTV value has been estimated with regression. In addition, by applying binning, it was turned into a classification problem and classification success was evaluated.
- Data Pre-Processing and First-Look
- Feature Extraction
- Analysis using PandasProfiler
- Hyper-Parameter Optimization
- Feature Selection
- Training and Prediction
- Unsupervised Clustering
- Binned Classification
- Conclusion
@misc{ozturk2021_revenue,
author = {Anil Ozturk},
title = {Estimating Revenue with in-app User Events},
year = {2021},
publisher = {GitHub},
journal = {GitHub repository},
howpublished = {\url{https://github.com/nlztrk/app_revenue_estimation}},
}