Microsoft Azure AI Fundamentals (AI-900) Practice Exam

Which technique involves dividing a date into separate components like month, day, and year?

1. Model evaluation

2. Feature selection

3. Feature engineering

4. Data preparation

The correct answer is: Feature engineering

The technique of dividing a date into separate components such as month, day, and year is known as feature engineering. This process involves transforming raw data into a format that better suits the needs of a machine learning model. By breaking down a date into individual components, you enhance the model’s ability to understand and utilize temporal information effectively. Feature engineering is crucial because it helps in creating new features that draw meaningful insights from the existing data. For example, treating the month, day, and year separately can allow the model to capture seasonal trends, patterns, or cyclic behaviors, which might not be possible if the date is kept as a single entity. In the context of the other choices, model evaluation relates to assessing the performance of a machine learning model after it has been trained, while feature selection involves choosing the most relevant features from a dataset for modeling, rather than creating new features. Data preparation encompasses the broader process of cleaning and organizing data before it is fed into a model, which may include initial steps similar to feature engineering but does not specifically refer to the transformation of date components.