Session 1 · How predictions get made · Lesson 9 of 36
Session 1 activities
ACTIVITIES · 28 MINFictional data only
All companies, people, and data are fictional. Never enter real customer, patient, or employee information (G.L. c. 93H, 201 CMR 17.00).
You can do every activity here. Sign in to save your responses and add your vote to the class tally.
Rule or learned pattern?
Sort each example.
Buckets: Rule written by a personLearned from data
- Flag every claim over $10,000
- Score claims by similarity to past confirmed fraud
- Send a reminder 48 hours before every appointment
- Predict which appointments are likely no-shows from lead time, weekday, and past attendance
- Forecast next month's demand for part BP-2207 from three years of orders
- Reorder when stock falls below 200 units
0 of 6 sorted
Feature, label, or output?
Clinic no-show model (fictional Charles River Family Dental).
Buckets: Feature (input)Label (past outcome)Output (prediction)
- Days between booking and appointment
- Whether the patient attended (past appointments)
- No-show risk score 0.62 for next Tuesday's 9 AM
- Day of week
- Reminder sent (yes/no)
- Number of prior missed appointments
0 of 6 sorted
A prediction I already rely on
Lab 1 warm-up.
Describe one prediction or score used in your workplace (CRM lead score, demand forecast, quality flag) in general terms. What goes in, what comes out, who acts, and what happens if it's wrong?
- No real customer or patient data.
- Name roles, not people.
- Say what a wrong prediction costs.