Removing incomplete or inaccurate records and removing columns that are not interesting are activities that would most likely take place in the __________ step of the CRISP-DM model.

• A. Data Preparation
• B. Evaluation
• C. Deployment
• D. Business Understanding

Answer: (A)

Data cleaning and feature selection are essential parts of preparing data for modeling. In CRISP-DM, Data Preparation follows Data Understanding and involves cleaning the data, handling missing values, correcting inaccuracies, and reducing dimensionality by removing columns that do not contribute to the analysis. Removing incomplete or inaccurate records and discarding uninteresting columns are classic preparation tasks because they improve data quality and reduce noise before building models. The other stages focus on defining business goals, examining the data in more depth, evaluating model performance, and deploying the model, not on cleaning and pruning data.