Design and validate a cost-sensitive classifier

Design a binary classifier that estimates the probability that a user will make a purchase within the next 7 days. The positive class is imbalanced: roughly 3% of examples are buyers. Labels are delayed after the prediction timestamp, with a purchase potentially taking up to 10 days to be confirmed. Available predictors include recent session-level statistics, count features, and category-level recency features. Assume the following business impact: a true positive generates…