Признаки из истории подписок для частично наблюдаемых пользователей
Признаки из истории подписок для частично наблюдаемых пользователей
Ответить самому
Сначала сформулируйте ответ как на собеседовании, затем откройте разбор и оцените себя.
Короткий ответ
Use RFM-style features plus subscription-state features: total paid, number of renewals, tenure, active status, days since last payment, payment regularity, plan price and normalized rates.
Полный разбор
For partially observed users, aggregate the transaction sequence as of the prediction timestamp. Useful features include total revenue so far, number of paid periods, tenure since acquisition, active subscription flag, current plan, last payment date, days since last payment, renewal streak length, gaps between payments and average monthly revenue so far.
The key is to preserve time shape, not only totals. Two users with the same number of payments are different if one paid recently and the other stopped months ago. Recency, active status, gaps and normalized features such as paid months divided by observed tenure help separate those cases.
Avoid leakage by computing all features only from events before the snapshot. Handle young users separately or include exposure time, because a user acquired yesterday has fewer possible renewals than a user observed for six months. For subscription LTV, survival-style features or hazard models can also be useful when churn timing matters.
Теория
A transaction history is a time series; flattening it to one lifetime sum loses recency and censoring information.
Типичные ошибки
- Use total payments only and ignore when those payments happened.
- Compare users with different observation windows without tenure features.
- Include future renewals in training features.
- Treat currently active and long-churned users with the same totals as identical.
Как отвечать на собеседовании
- Say “recency, frequency, monetary value, tenure and active state”.
- Call out snapshot-time feature generation.