This project replaces traditional, static audit schedules with a Predictive Oversight model. By utilizing Random Forest machine learning, the analysis identifies the mathematical "Risk Signatures" that actually lead to failures—specifically isolating Systemic Risk Factors as the primary driver over simple asset value. The resulting framework segments operations into three strategic tiers, allowing the organization to stop wasting resources on stable units and instead concentrate 70% of audit capacity on high-risk clusters where failure is statistically most likely to escalate.

La préparation des données du hackaviz 2026

Ejercicio de Prueba de Normalidad

La Ordenación de burbuja (Bubble Sort en inglés) es un sencillo algoritmo de ordenamiento.

Pruebas de normalidad

This lab explores multiple logistic regression using a sample of 5,000 U.S. adults from the 2020 Behavioral Risk Factor Surveillance System (BRFSS). The outcome of interest is frequent mental distress (14+ mentally unhealthy days in the past 30). The analysis covers building and interpreting an adjusted logistic regression model, conducting Wald and likelihood ratio tests, testing an interaction between smoking status and sex, and evaluating model performance through McFadden's pseudo-R², the Hosmer-Lemeshow test, a calibration plot, and an ROC curve (AUC = 0.736).