Case Study

Assess a source tag for resolving conflicting labels in a two-region pricing model.

Case context: You are building a model to predict the sale price of used bicycles. Your training data comes from two regions, Coastal and Inland, where similar bicycles often sell for very different amounts because local demand and shipping costs differ. You are considering adding a feature that identifies the region for each training example.

Question: Based on the underlying principle, explain what this region tag does to the target value y and whether this is a common practice in machine learning.

Sample answer: Adding the region tag to the input x makes the target y better defined, because the same bicycle can map to different prices depending on whether it comes from Coastal or Inland. This removes the ambiguity in the training examples. Even so, this kind of source-indicator feature is not a standard solution that people use very often in practice.

Key points:

  • A region tag in x makes y less ambiguous.
  • The feature helps separate examples from Coastal and Inland.
  • The idea is clear and valid, but it is not widely used as a routine fix.

Rubric: The evaluation must say that specifying the region in input x makes the target value unambiguous and note that this approach is not commonly used in practice.

0

1

Updated 2026-08-12

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy

Machine Learning Yearning @ DeepLearning.AI

Related