Hira Irfan is a Master of Engineering student in Data Analytics and Machine Learning in the Edward S. Rogers Sr. Department of Electrical & Computer Engineering at the University of Toronto. She holds a Bachelor of Computer Science from Toronto Metropolitan University.

Her research focuses on explainable artificial intelligence for cybersecurity, with an emphasis on interpreting deep learning-based anomaly detection models for network flow data. She studies autoencoder-based approaches for modeling normal network behavior and detecting anomalous traffic through reconstruction patterns. Her work applies feature-level reconstruction analysis, SHAP-based attribution methods, feature interaction analysis, and perturbation-based faithfulness evaluation to better understand why models flag traffic as anomalous.

Hira’s broader interests include machine learning, cybersecurity, model interpretability, data analytics, and trustworthy AI systems. Her research aims to make AI-driven threat detection more transparent, reliable, and useful for security analysts.