Answer: Precision

## Detailed Explanation In fraud detection systems, the primary business objective often involves optimizing resource allocation for manual review. The scenario describes a financial institution aiming to **minimize employee time spent reviewing cases that are ultimately non-fraudulent** (false positives). This directly aligns with optimizing **Precision**. ### Why Precision (Option C) is the Correct Metric **Precision** is defined as: \[ \text{Precision} = \frac{\text{True Positives}}{\text{True Positives} + \text{False Positives}} \] - **True Positives (TP)**: Cases correctly flagged as fraud. - **False Positives (FP)**: Cases incorrectly flagged as fraud (non-fraudulent cases sent for review). A high precision value indicates that when the system flags a transaction as fraudulent, it is highly likely to be actual fraud. This minimizes the number of false positives that employees must investigate, thereby reducing wasted review time on legitimate transactions. ### Analysis of Other Options - **A: Recall** - Measures the proportion of actual fraud cases that are correctly identified (True Positives / [True Positives + False Negatives]). While important for catching fraud, optimizing recall might increase false positives, leading to more non-fraudulent cases being reviewed—contrary to the stated goal. - **B: Accuracy** - Measures overall correctness (True Positives + True Negatives) / Total Cases. In imbalanced datasets like fraud detection (where fraud cases are rare), accuracy can be misleadingly high even if the model performs poorly on the minority class. It doesn't specifically address the reduction of false positives. - **D: Lift Chart** - A visualization tool used in marketing and campaign response modeling to compare model performance against random selection. While useful for certain analyses, it is not a direct evaluation metric for minimizing false positives in fraud detection scenarios. ### Business Context Alignment In practical fraud detection implementations, precision is prioritized when the cost of investigating false positives (employee time, customer inconvenience) is high. The financial institution's explicit goal of reducing time spent on non-fraudulent reviews makes precision the most appropriate metric to monitor and optimize during model development and deployment.

Author: LeetQuiz Editorial Team