CPMAI's Task: AI Pattern Identification requires teams to map their specific business objectives to the most appropriate one or more of the Seven Patterns of AI. Starting from "what are we trying to accomplish?" and then selecting the pattern(s) that align with those goals is the prescribed approach.
=========

CPMAI's Generic Task Group: Select Modeling Technique in Phase IV: Model Development outlines common cognitive algorithms. For recommendation systems-which rely on finding similar user or item profiles-the K-Nearest Neighbor algorithm is the canonical choice, using customer purchase vectors to locate "nearest neighbors." In contrast, K-means is purely unsupervised clustering, Neural Networks are more complex and not necessary for basic collaborative filtering, and Hyperpersonalization is an AI pattern, not an algorithm.
=========

Trustworthy AI is not confined to a single phase but is woven throughout the entire CPMAI lifecycle:
The CPMAI Exam Content Outline under Domain VI: Trustworthy AI specifies tasks such as "Apply ethical AI concepts throughout the development lifecycle," "Ensure compliance with privacy/security requirements," and "Implement transparency and explainability" at every stage .
The CPMAI Workbook's Task Group: Trustworthy AI Requirements (covering transparency, explainability, ethics, compliance, and responsible-AI frameworks) appears as an overarching set of artifacts and considerations that map back to multiple phases-beginning with Business Understanding and continuing through Model Operationalization .
Thus, Trustworthy AI considerations apply across all CPMAI phases.

A false positive occurs when the model predicts the positive class (e.g., "tree") but the actual label is negative (e.g., "flower"). The confusion matrix definition confirms that mislabeling a negative instance as positive maps to the false positive count.
=========