Intermediate level · Module 6: Intermediate Machine Learning Project · Lesson 1 of 3
Framing the Project
Define the decision, the success criteria and the data available before building anything.
In this lesson you will
- Write a project brief with decision, users and success criteria
- Identify data, constraints and risks
Key concepts
project briefsuccess criteriaconstraintsrisks
The lesson
State the decision: which agents should receive extra cash next week? Name the users — operations staff — and how they will use the prediction.
Success criteria combine model metrics and business outcomes: catch at least 70% of shortages with no more than two unnecessary deliveries per correct one, and beat the current rule of thumb.
List the data and its limits, constraints such as weekly retraining on a laptop, and risks such as unfair treatment of agents in remote areas.
Step by step
- Write the decision and users.
- Set measurable success criteria.
- List data, constraints and risks.
- Agree the brief with the instructor.
Practice activity
Submit the project brief.
Expected result: An approved brief with measurable, business-relevant criteria.
Check your understanding
-
Why include the current rule of thumb in success criteria?
Show answer
The model must beat what the business already does to be worth using.
-
Name one constraint that affects the design.
Show answer
For example retraining weekly on modest hardware.
Taught live by an instructor — online or in person. Free course lessons are for self-study and do not lead to a certificate — see registration options.