What technique is commonly used to predict the price of a house based on its features?

The technique commonly used to predict the price of a house based on its features is regression. This method is appropriate because it deals with continuous dependent variables; in this case, the house price is a continuous value. Regression models can analyze the relationship between the house's various features—such as square footage, number of bedrooms, location, and amenities—and predict a numerical outcome, which is the price.

In contrast, classification is used for categorical outcomes, where the goal is to predict discrete labels rather than continuous numerical values. Clustering involves grouping data points based on similarities in their features but does not directly predict an outcome. Association rules focus on discovering interesting relationships or patterns within datasets, but again do not provide a prediction mechanism for a variable like house price. Hence, regression is the most fitting technique for this predictive task.