What type of machine learning algorithm would you use to predict the resale price of a residential property?

The prediction of resale prices for residential properties is a classic example of a problem that requires regression analysis. Regression algorithms are specifically designed to establish a relationship between a dependent variable (in this case, the resale price) and one or more independent variables (such as property size, location, number of bedrooms, and other relevant features).

By employing a regression model, you can predict continuous values, which is essential for accurately forecasting prices that can fall anywhere on a numeric scale. These models can help identify trends and patterns in historical property sales, making them valuable for understanding how various factors influence pricing.

Classification, on the other hand, is used for problems where the output is a category or class label (e.g., whether a property will be sold above or below a certain price threshold), which doesn't apply to predicting continuous values like resale prices. Clustering focuses on grouping data based on similarities without predicting a specific outcome, suitable for exploratory analysis rather than direct predictions. Dimensionality reduction aims to simplify datasets by reducing the number of variables while retaining essential information, but it is not directly involved in price prediction tasks.

Therefore, using regression for predicting the resale price is the most appropriate choice, as it effectively handles continuous numerical outputs.