Which term refers to preparing data for modeling by addressing missing values and inconsistencies?

Which term refers to preparing data for modeling by addressing missing values and inconsistencies?

• A. Data cleansing and transformation
• B. Data augmentation
• C. Data acquisition
• D. Feature engineering

Answer: (A)

Preparing data for modeling involves cleaning and transforming data so models can learn reliably. Addressing missing values and inconsistencies is the essence of data cleansing, which means detecting and handling incomplete or erroneous data. Transformation adds the steps of converting data into a uniform format, scaling or normalizing features, and encoding categories so all data align well with modeling algorithms. This combination ensures high data quality and consistency, which directly strengthens model performance and training stability. Data augmentation, on the other hand, is about creating extra synthetic data to expand the dataset; data acquisition is about collecting data; feature engineering focuses on creating new features to improve models, not primarily on cleaning and standardizing existing data.

Preparing data for modeling involves cleaning and transforming data so models can learn reliably. Addressing missing values and inconsistencies is the essence of data cleansing, which means detecting and handling incomplete or erroneous data. Transformation adds the steps of converting data into a uniform format, scaling or normalizing features, and encoding categories so all data align well with modeling algorithms. This combination ensures high data quality and consistency, which directly strengthens model performance and training stability. Data augmentation, on the other hand, is about creating extra synthetic data to expand the dataset; data acquisition is about collecting data; feature engineering focuses on creating new features to improve models, not primarily on cleaning and standardizing existing data.