Engineer features with a source, unit and window
A feature is useful only when you can explain how it was made, what it measures, and whether it was available in time.
The sum of W1-W4 logins [3,4,2,5] is 14 logins, usable only after all four weeks are complete and accessible. It is not hours, attendance or direct proof of engagement.
Minutes per login: R has 200/10=20; S has 100/5=20. The same ratio conceals different totals. Keep totals if they matter. With zero logins, the ratio is undefined; use missing/None and a zero-login indicator. Investigate a combination such as positive minutes and zero logins according to source definitions instead of assuming a correction.
Elapsed time uses the correct event: at the 5 September 2026 snapshot, first registration on 1 September gives 4 days and on 4 September gives 1 day. A record-update timestamp is not necessarily first registration. Use documented formats and time zones where they affect the calculation.
Sources: W4 PDF pp. 2-42; L4 Tasks 1-7. All fixed worked values here were recalculated from the supplied examples. These toy rules are teaching devices, not deployable student-support recommendations.
The supplied split tells you which records teach the rules and which records only receive predictions. Keep those roles separate.
Practice problems
Supplementary study questions on the lecture concepts — not verbatim exam questions. Each answer counts toward spaced review.
1. Conceptual understanding
Definitions and distinctions.
What must an engineered feature document?
How should minutes per login handle zero logins?
2. Application
Apply the concept to a case or calculation.
Which total matches early_logins for an end-of-week-4 task?
A student has 120 minutes and 6 logins. What is minutes per login?
3. Interpretation and limitations
Evaluate assumptions and justify conclusions.
A student has zero logins. What should happen to minutes/login?
A ratio uses imputed minutes. Which description is accurate?
Explain it in writing
Self-checked, not marked. Write a few sentences, then compare with the example answer. Your writing is saved as you type.
Show an example answer
Its source, meaning/unit, time window and availability, plus rules for missing inputs or exceptional cases such as zero denominators.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?
Show an example answer
The ratio depends partly on a replacement assumption. Marking this preserves the distinction between a fully observed measurement and one based on an estimate.
Check: did you answer the question, use evidence or working, and avoid claiming more than the evidence supports?