Microsoft Azure AI Fundamentals (AI-900) Practice Exam

What type of models are typically used to predict sale prices of auctioned items?

1. Classification models

2. Regression models

3. Clustering models

4. Decision tree models

The correct answer is: Regression models

Regression models are specifically designed for predicting continuous numerical values, making them ideal for forecasting sale prices of auctioned items. In this context, the objective is to estimate a specific price based on various input features, such as item characteristics, historical sale prices, demand trends, and more. By utilizing regression analysis, the model can uncover relationships between these features and the predicted sale price, allowing it to make informed predictions that reflect the potential value in an auction scenario. This type of modeling accounts for variability in pricing and can handle the nuances of several influencing factors that might affect the price of auctioned items. While decision tree models can be considered a form of regression model, especially when they're used for predicting continuous variables, they are not categorized separately in this context. Classification models, on the other hand, focus on categorizing data into distinct groups and are not suitable for predicting continuous values like sale prices. Clustering models aim to group similar items together rather than predict specific outcomes based on input data, which further distinguishes them from regression models in the scenario described.