Session 1 · AI and ML fundamentals (Domain 1) · Lesson 9 of 37
Session 1 activities
ACTIVITIES · 31 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.
Supervised, unsupervised, or reinforcement?
Sort each fictional business problem.
Buckets: SupervisedUnsupervisedReinforcement
- Predict whether a claim is fraud from 5 years of labeled claims
- Forecast next week's room-nights from past bookings
- Group machines by similar vibration patterns, no labels
- Flag unusual sensor readings with no failure labels
- Train a warehouse robot by rewarding efficient picking routes
- Predict a patient's appointment no-show from labeled history
0 of 6 sorted
Read the confusion matrix
Branching. Pioneer Valley Mutual (fictional) tests a fraud model on 1,000 claims, 50 of them fraud. Results: true positives 30, false negatives 20, false positives 10, true negatives 940.
What is the precision?
Is ML the answer?
Poll, then discuss.
Charles River Family Dental wants to send a reminder 48 hours before every appointment. What is the right tool?
My study plan (W1)
Online, after session 1.
Write your W1 study plan: exam date target, weekly hours, your weakest domain from the diagnostic, and the two resources you will use.
- A date.
- Hours per week.
- Weakest domain and why.
- Two resources (exam guide plus one more).