Case Study

Categorize the spam data collection strategy during an anti-spam system kickoff.

Case context: A machine learning engineering team is tasked with building a new email anti-spam system. During the initial brainstorming session, the team generates a list of potential project paths. One of the proposed paths is to collect a huge training set of spam email. A debate arises over whether the team should immediately commit to this path or treat it as one of several options.

Question: Based on the course concepts, how should the team classify this proposed path, and what is the relationship between this proposal and the other ideas generated during kickoff?

Sample answer: The team should classify the proposal to collect a huge training set of spam email as one possible direction for building the anti-spam system. According to the course context, a team typically has several ideas when starting out. Therefore, this proposal is one of multiple potential directions and should be evaluated against other brainstormed options rather than being adopted immediately as the sole project plan.

Key points:

  • Classify the proposal as one possible direction for building the anti-spam system.
  • Acknowledge that the team has several brainstormed ideas at the beginning of the project.
  • Evaluate this direction as one option among a broader set of potential paths.

Rubric: Students must identify that collecting a huge training set of spam email is classified as one possible direction (or path) for the project. They must state that it is one of several brainstormed ideas that should be compared and evaluated rather than being treated as the only choice.

0

1

Updated 2026-05-27

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Machine Learning Strategy

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Yearning @ DeepLearning.AI

Related