Hadi Jahanshahi

/hɒdi jæhɒnʃɒhi/ (Lead Data Scientist, Machine Learning Engineer)

Hadi Jahanshahi received his Ph.D. degree from the Data Science Lab (DSL) at Toronto Metropolitan University (Formerly Ryerson University), Toronto, Canada. He was an ML researcher at the DSL between 2018 to 2022. He received his M.Sc. from the Department of Industrial Engineering at Boğaziçi University, Istanbul, Turkey, in 2018, and his B.Sc. degree from the Department of Industrial Engineering at Iran University of Science and Technology Tehran, Iran. His research interests are Machine Learning, Reinforcement Learning, Sequence Mining, Preference Learning, Software Engineering, and Behavior Analysis.

His Ph.D. dissertation was on AI models to automate bug triage in software issue tracking systems. To address the problem, he utilizes Natural Language Processing, Machine Learning, Reinforcement Learning, Approximate Dynamic Programming, Simulation, Graph Analysis, and Optimization (Integer Programming). Most of the codes related to the adopted methods are publicly available on his GitHub page.

Hadi is currently working at the Royal Bank of Canada as a lead data scientist. As an ML member of the Next Best Action team, he explores state-of-the-art ML algorithms to recommend the most appropriate products to clients.

Recent Published Works (Data Science Related)

ADPTriage: Approximate Dynamic Programming for Bug Triage

Jahanshahi, H., Cevik, M., Mousavi, K., & Başar, A. (2023). ADPTriage: Approximate Dynamic Programming for Bug Triage. IEEE Transactions on Software Engineering, 4594-4609, v. 49. [download][codes][preprint]

A deep reinforcement learning approach for the meal delivery problem

Jahanshahi, H., Bozanta, A., Cevik, M., Kavuk, E. M., Tosun, A., Sonuc, S. B., ... & Başar, A. (2022). A deep reinforcement learning approach for the meal delivery problem. Knowledge-Based Systems, 243, 108489. [download][codes][preprint]

nTreeClus: A tree-based sequence encoder for clustering categorical series

Jahanshahi, H., & Baydogan, M. G. (2022). nTreeClus: A tree-based sequence encoder for clustering categorical series. Neurocomputing, 494, 224-241. [download][codes][preprint]

Honors & Awards

Natural Sciences and Engineering Research Council (NSERC)

My Ph.D. dissertation was in part supported by NSERC Discovery Grant. I proposed a recommendation system for the bug triage in open-source software repositories. I leveraged Simulation, Machine Learning, Natural Language Processing, Markov Decision Process, Integer Programming, and Approximate Dynamic Programming to address the problem. The grant was given to my supervisor professor Ayşe Başar.

MIE Graduate Scholarship top-up

The Department of Mechanical and Industrial Engineering, TMU is providing this funding to assist active research students in the graduate program. I have been awarded for the winter 2021 term.

Ryerson International Student Scholarship (RISS)

The RISS is an award in recognition of the high academic standing and is awarded to students on a competitive basis. These awards are distributed through the doctoral programs in engineering.

Mitacs Accelerate Program (Fund)

We had a project with "Your Doctors Online" company, which was funded and supported by Mitacs through the Mitacs Accelerate Program. We proposed a Smart Reply system to help physicians chat with patients. Our paper was published in the Journal of Healthcare Informatics Research. The grant was given to my supervisor professor Mucahit Cevik.

Air Force Office of Scientific Research Grant

Secured the Air Force Office of Scientific Research Grant for my master's thesis under my supervisor's guidance, wherein we developed the 'nTreeClus' model, a tree-based sequence encoder designed for clustering categorical series, focused on the niche area of unevenly spaced multivariate time series with mixed variable types.

Ryerson Graduate Fellowships (RGF)

A Ryerson Graduate Fellowship is a merit-based scholarship for students pursuing studies in one of the university’s graduate programs. The allocation of fellowships is on a competitive basis and is consistent with Ryerson’s policies concerning access, equity and research integrity, and the criteria established by the Scholarship and Awards Committee of the Yeates School of Graduate Studies. The Committee oversees the evaluation of candidates and selection of award winners in collaboration with Graduate Program Scholarship Committees. In my case, it is renewed for three years.