Microsoft Azure AI Fundamentals (AI-900) Practice Exam

Which method would you use to select the most relevant variables for a predictive model?

1. Feature engineering

2. Feature selection

3. Model optimization

4. Data augmentation

The correct answer is: Feature selection

Feature selection is the process used to identify and select the most relevant variables for a predictive model. This method helps to improve the model's performance by eliminating irrelevant or redundant features that do not contribute significantly to the prediction. By focusing on the most impactful variables, the model can become more efficient, potentially leading to better accuracy and less overfitting. Feature selection techniques often assess the relationships between predictor variables and the target variable. This may include statistical tests, recursive feature elimination, or algorithms that inherently perform feature selection, such as tree-based models. The goal is to simplify the model without losing predictive power, making it easier to interpret and faster to run. On the other hand, feature engineering involves creating new features or modifying existing ones to improve model performance. This can be useful but does not specifically address the direct selection of relevant features. Model optimization is related to tuning hyperparameters and improving a model's performance but also does not focus on choosing the input variables. Lastly, data augmentation refers to techniques that increase the diversity of the training dataset by applying transformations, and it does not pertain to the selection of relevant features. Selecting the right features is crucial in the model-building process, as using irrelevant features can lead to a decrease in model performance, while the