In the context of preparing data for machine learning models, data transformation plays a pivotal role. A team is working on a project that involves predicting customer churn for a telecom company. The dataset includes customer demographics, service usage, and complaint history. The raw data is messy, with missing values, inconsistent formats, and categorical variables not suitable for direct input into machine learning algorithms. The team needs to preprocess this data to make it suitable for analysis. Which of the following best describes the process of 'data transformation' in this scenario? Choose the best option.