Insurance fraud detection with process mining
Three-layer architecture and the FWE metric for detecting fraud from process trajectories.
IN DEVELOPMENTDrag the sphere or pick a node to see its skills and public repository.
I work at the intersection of artificial intelligence, statistics and process analysis. I'm interested in systems where anomalies are systemic: they show up in the data and in how processes relate to each other.
I'm a Data Quality Analyst at Shift Technology, an InsurTech company. I'm pursuing a Master's in Artificial Intelligence Engineering (Big Data) at the Universidad Politécnica Metropolitana de Hidalgo.
Working principle: no result is presented without a confidence interval, a statistical test and explicit limitations.
Shift Technology · InsurTech. Data validation and traceability for insurance systems.
Aeroméxico. Sales analysis to identify improvement areas and optimize commercial processes.
Three-layer architecture and the FWE metric for detecting fraud from process trajectories.
IN DEVELOPMENTLBP + SVM vs. ResNet-18, with 95% confidence intervals and McNemar's test.
COMPLETEDClassical pipeline with Lab chroma, Otsu thresholding and threshold sensitivity analysis.
COMPLETEDComputer vision practices: filters, pothole detection and steel defects, with reproducible code.
IN PROGRESSEducation
Certifications
Let's talk about data quality, computer vision and process mining.