In Phase III: Data Preparation, the Select Data task instructs teams to choose only the records and attributes needed for modeling—documenting inclusions and exclusions to reduce volume and complexity. This selective pruning of columns and rows is the primary mechanism for trimming excessive data before modeling.

=========

The CPMAI Glossary defines a naive Bayes classifier as “a family of simple probabilistic classifiers based on Bayes’ theorem with the assumption of feature independence,” making it ideal for small or limited datasets where model simplicity and interpretability are priorities.

=========

Under Domain VI: Trustworthy AI in the CPMAI Exam Content Outline, Responsible AI covers establishing policies, guidelines, and governance that ensure AI solutions are developed for positive, ethical use and prevent misuse. Defining proper-use guidelines and embedding ethical intent into facial recognition directly align with Responsible AI practices .

In Phase II: Data Understanding, CPMAI’s Data Selection tasks require teams to “Decide on the data to be used for analysis” by first listing all available sources and then selecting only those records and attributes that meet quality and relevance criteria . Taking a company-wide data inventory ensures you don’t overlook relevant datasets before narrowing down to what truly applies.

=========

The CPMAI™ v7 methodology groups all data-centric preparation activities—including both data cleansing (“Clean data”) and data augmentation (“Enhance & Augment data”)—into Phase III: Data Preparation. In this phase, teams focus squarely on constructing the dataset to be used for modeling by performing all required cleaning, transformation, and enhancement operations.

Phase III: Data Preparation is defined in the Workbook’s Table of Contents as covering Data Cleansing & Enhancement tasks (“Clean data” and “Enhance & Augment data”) .

Under Phase III, the Generic Task Group: Data Cleansing & Enhancement explicitly lists “Task: Clean data” (bringing data quality to modeling-ready levels) and “Task: Enhance & Augment data” (producing derived attributes and new records) as core activities .

=========

Generative AI is defined in the CPMAI Glossary as “AI systems that create new data (e.g., text, images, music) based on patterns learned from existing data.” Using Generative AI for content generation directly addresses the need to produce marketing images automatically.

=========

In Phase II: Data Understanding, CPMAI’s Generic Task Groups focus on:

Collecting initial data (identifying sources and volumes) and describing data (location/source) .

Verifying data quality to ensure completeness and correctness .

Team composition (the makeup of your data team) is addressed earlier under Phase I: Assess Situation, not during the Data Understanding phase where data requirements (quality, volume, source) are determined.

=========

Domain IV: Data for AI – Task 2: Implementing Data Governance and Management mandates establishing data stewardship, management plans, lineage, and master-data practices. Core governance concerns include how data is shared (A), ensuring user privacy (B), guarding against breaches (E), and maintaining high data quality (F). Quantity or change-management issues are operational rather than governance controls.

=========

The CPMAI™ Glossary contrasts automation (for deterministic, rule-based tasks) with AI (for probabilistic, learning-based tasks). As a rule of thumb, if a problem exhibits probabilistic patterns that can’t be captured by fixed rules, then AI is the appropriate solution; deterministic problems are better handled by simple automation.

CPMAI defines a Data Scientist as the role responsible for “formulating data-driven hypotheses, selecting and applying statistical algorithms, interpreting model results, and communicating insights to stakeholders,” all of which require critical thinking under uncertainty, advanced mathematics, and strong data-visualization skills .

=========