What is the technique called that improves a model’s ability to make estimations by generating features?

The technique that enhances a model's ability to make estimations by generating new features is known as feature engineering. This process involves using domain knowledge to create features that better represent the underlying problem to the predictive models, thus improving their performance.

In feature engineering, practitioners create new features through various methods such as combining existing features, creating interaction features, or applying mathematical transformations. The goal is to provide the model with more informative inputs that can capture the complexities of the data, ultimately leading to more accurate predictions.

Feature extraction, while related, typically refers to the process of reducing the dimensionality of data by transforming it into a set of features that summarize the original data. This is often used in contexts such as image processing or natural language processing.

Feature selection involves choosing a subset of relevant features from the entire set, instead of creating new ones. This can improve model performance by eliminating irrelevant or redundant features but does not enhance the feature set itself.

Data transformation generally refers to modifying the data into a different format or structure, which can include normalization or scaling, but does not specifically focus on feature creation.

In summary, feature engineering is the technique that involves generating new features to improve model estimations, making it the correct choice in this context.