The core principle defines the True Positive Rate (TPR), formally known as sensitivity or recall, as a fundamental metric in binary classification theory that quantifies the probability of correctly identifying positive instances within a target population. The concept relies on formal set-theoretic definitions distinguishing true positives from false negatives under strict probabilistic constraints inherent to statistical decision theory. This measure functions as an essential component of the operating characteristic space within pattern recognition and machine learning, providing a boundary condition for evaluating classifier performance independent of class prevalence.
Prerequisite chain context: requires Threshold Selection Strategies for Models.