About
About the Lab
Who we are
The Data Lab is a research group in the Department of Electrical, Computer and Biomedical Engineering at Toronto Metropolitan University, directed by Dr. Faezeh Ensan. We study how machines can understand, retrieve and reason over text: building knowledge graphs and the models that reason over them, designing neural search and ranking systems, automating the search for evidence in the biomedical literature, and making large language models fairer and more reliable.
Much of our work sits where structured knowledge meets natural language. We are equally interested in the practical question of whether a method holds up outside the lab, which is why we build and release benchmarks and datasets, evaluate on real search systems such as PubMed, and prefer open, reproducible models where they can match closed ones.
The lab has trained PhD, Master's and undergraduate students who have gone on to positions at Google, Amazon, AMD, Electronic Arts, Thomson Reuters, the Vector Institute and graduate programs across North America. Our research has appeared in venues including ACM Transactions on Information Systems, Information Processing & Management, Knowledge and Information Systems, Machine Learning, the Journal of Biomedical Informatics, SIGIR, CIKM and WSDM.
Our history
- 2006–2011
Roots in semantic technologies
Dr. Ensan's doctoral work at the University of New Brunswick on modular ontologies and knowledge encapsulation earns the International Semantic Web Student Prize (ISWC 2008) and the NeOn best-paper prize (EKAW 2008).
- 2011–2018
From semantics to search
Postdoctoral and industry research at UBC, Athabasca University and SideBuy Technology shifts the focus toward entity linking, semantic search and neural ad-hoc retrieval. Dr. Ensan leads a team to a top-four finish among university groups at the TREC 2016 Contextual Suggestion track, and supervises her first graduate students at Ferdowsi University of Mashhad.
- 2019
Data Lab is founded
Dr. Ensan joins the Department of Electrical, Computer and Biomedical Engineering at Toronto Metropolitan University and establishes the Data Lab, with an initial focus on knowledge-graph-based retrieval and query performance prediction.
- 2020–2022
Biomedical literature search
The lab turns to a pressing real-world problem: helping researchers find evidence in the flood of biomedical publications. Work on query expansion for COVID-19 scholarly search and technology-assisted reviews appears in the Journal of Biomedical Informatics and at Canadian AI, and the lab co-chairs the SeBiLAn 2022 workshop on semantics-enabled biomedical literature analytics. Papers on entity retrieval and learning to rank are published at SIGIR and in Information Processing & Management.
- 2023–2024
Question answering and responsible AI
The lab's first PhD graduate, Parastoo Jafarzadeh, completes a thesis on leveraging textual context in knowledge graphs. Research broadens to knowledge-graph question answering, SPARQL generation with generative language models, and fully automated scholarly search (the FASS-BSLR benchmark). Dr. Ensan receives the Dean's Teaching Award and co-edits a Journal of Biomedical Informatics special issue.
- 2025–present
Trustworthy language models
Recent work tackles bias mitigation in transformers (Schema-Tune), ambiguity and unanswerability in question answering, and open, medium-sized generative models for biomedical systematic reviews, published in ACM TOIS, Machine Learning, Artificial Intelligence Review and at SIGIR-AP and ICTIR. Dr. Ensan is promoted to Associate Professor in 2026 and joins the editorial board of Discover Artificial Intelligence.