Shahrzad Presents Research on Sparse Oblique Rule Boosting at ECML PKDD 2026
Huge congratulations to Shahrzad and the team on having their paper accepted into the journal Data Mining and Knowledge Discovery and presented as part of the NECTAR track at ECML PKDD 2026 in Naples, Italy! 
Her research introduces sparse oblique rule boosting, a method designed to learn interpretable rule ensembles using sparse oblique conditions.
Why it matters: A major challenge in machine learning models is balancing predictive performance with human interpretability. By introducing linear representation learning into rule ensemble models, Shahrzad’s work significantly improves the simplicity-accuracy trade-off, making models clearer and easier to interpret without sacrificing predictive power.
Check out the full journal publication or reach out to Shahrzad to learn more about the methodology!
Please join us in celebrating Shahrzad for this well-deserved recognition in top-tier data science!
