/fintech-user-engagement

Defining, predicting, and preventing dropout users in fintech setting

Primary LanguageJupyter Notebook

Defining, Predicting, and Preventing Disengaged Users in FinTech

Designing and developing machine learning models and AB testing experiments to reduce disengaged users in FinTech

Part I:

  1. Defining engagement
  2. Predicting engagement (data pipeline, feature extraction, and model training using pySpark and Scikit-Learn)
  3. Model evaluation

Part II:
4. Designing intervention (contextual recommender system using ALS)
5. Designing A/B Testing
6. Evaluating and interpreting the results

Article published in: https://towardsdatascience.com/defining-predicting-and-preventing-disengaged-users-in-fintech-30dcb3bc0460

Singapore, May 2021