Microsoft Azure AI Fundamentals (AI-900) Practice Exam

Which process is used to create additional features from raw data in AI development?

1. Data mining

2. Feature engineering

3. Data normalization

4. Model training

The correct answer is: Feature engineering

Feature engineering is the process of creating additional features from raw data in AI development. This involves transforming and enhancing the dataset to improve the performance of machine learning models. By selecting, modifying, or generating new features, practitioners can better capture the underlying patterns in the data, leading to more robust and accurate models. Effective feature engineering often includes techniques like encoding categorical variables, creating interaction terms, or deriving new features based on existing data, all of which help to provide machine learning algorithms with the necessary information for learning. While data mining refers to discovering patterns in large datasets, it does not specifically focus on creating features. Data normalization relates to adjusting the range of the data to improve model training and performance but does not involve the actual creation of new features. Model training is the process where a machine learning algorithm learns from the processed data but does not pertain to the feature creation stage itself.