AIF-C01 Practice Question 31: Regression Numeric Prediction
AWS Certified AI Practitioner (AIF-C01) · Test 4 · question 1 of 10
A real estate company has years of records of past home sales. Each record includes the number of bedrooms and bathrooms, the square footage, the neighborhood, and the final sale price. The company wants a model that estimates the sale price, in dollars, of homes that come onto the market. Which ML technique fits this problem?
Show answer and explanation
Answer:
This is a supervised learning problem, because every historical record includes the target value, the sale price. When the target is mathematically continuous, such as a price in dollars, the problem is regression. Regression estimates a dependent target value from attributes that are correlated with it, and AWS uses exactly this example: predicting house prices from bedrooms, bathrooms and square footage (option B).
- A is wrong because binary classification assigns each item to one of two classes, like "sells" or "doesn't sell". It can't produce a dollar amount. - C is wrong because clustering is *unsupervised*: it finds groupings in data that has no target values. It would ignore the labeled prices the company already has and wouldn't output a price. - D is wrong because multiclass classification predicts a *category*. Bucketing prices into bands throws away the precise numeric estimate the company asked for.
Reference: Types of Algorithms