Advanced Techniques in Static Analysis for Performance Bug Prediction (1st Offering) - JLR Challenge # 4Technical Workshop Series by: Wardah Saleh

Monday, November 3, 2025 - 10:00
School of Computer Science – JLR Challenge # 4Technical Workshop Series

 

Advanced Techniques in Static Analysis for Performance Bug Prediction (1st Offering)

Presenter: Wardah Saleh

 

Date: Monday, November 3, 2025

Time: 10.00 am

Location: Workshop Space, 4th Floor - 300 Ouellette Ave., School of Computer Science, Advanced Computing Hub

 

Abstract

This workshop provides a comprehensive exploration of advanced techniques that enhance static analysis for predicting and preventing performance issues in large-scale software systems. Building on foundational concepts, it explores how machine learning, automation, and semantic code analysis are transforming static analysis into an intelligent, proactive process. Participants will learn about enhanced SZZ labelling, AutoML-based feature engineering, and advanced models such as LightGBM, CodeBERT, and Graph Neural Networks (GNN)) for understanding complex code behaviour. The session also highlights function-level precision, cross-repository learning, and developer metrics that capture the socio-technical aspects of bug prediction. Emphasizing interpretability, CI/CD integration, and continuous performance assurance, this workshop equips attendees with the knowledge to build scalable, explainable, and data-driven systems for early detection and prevention of performance issues — moving from detection to prediction to intelligent automation.

 

 

 

Workshop Outline:

1. Advanced static analysis (AST, CFG, data flow)

2. Enhanced SZZ labeling and data accuracy

3. AutoML-based feature engineering and optimization

4. Advanced ML models: LightGBM, CodeBERT, GNNs

5. Temporal and developer metrics integration

6. Function-level precision and cross-repository learning

7. Model interpretability (SHAP, LIME)

8. CI/CD automation and industrial applications

9. Continuous, explainable, adaptive performance assurance

 

Prerequisites:

Basic understanding of static analysis and familiarity with machine learning concepts or software development tools.

 

Biography

Wardah Saleh is currently a Ph.D. student in the School of Computer Science at the University of Windsor and an Assistant Professor in the Department of Computer Science at the American International University-Bangladesh (AIUB), where she is on study leave. She earned both her B.Sc. in Computer Science & Engineering and M.Sc. in Computer Science (Computer Network and Architecture) from AIUB, graduating with highest academic distinctions. Her research interests include VANET (Vehicular Ad hoc Networks), 5G technologies, IoT (Internet of Things), network security and AI (Artificial Intelligence). She also holds professional certifications such as CCNA (Cisco Certified Network Associate) and CCNP (Cisco Certified Network Professional).

 

Registration Link ( Only MAC need to pre-register)