Research
Our research brings semantics, computational modeling, and artificial intelligence together to build systems that can represent complex domains, reason across data sources, and help people evaluate choices and consequences.
Semantic Intelligence and Interoperability
We study how concepts, relationships, and context can be represented explicitly, allowing data from different organizations and information systems to be interpreted and used together. The work includes ontology design, semantic alignment, knowledge graphs, provenance, and cross-domain data standards.
Related projects: Core Domain Ontology, Common Impact Data Standard, Ontology of Social Service Needs, and Service Ontology.
Hybrid AI and Knowledge-Based Reasoning
We combine structured knowledge and symbolic techniques with machine learning and natural language processing. Research topics include explainable information extraction, graph-based retrieval and reasoning, and methods to validate information before it is used in decision support.
Related projects: SMILE and PARLANCE.
Computational Modeling of Social Systems
Our work develops models of people, goals, preferences, and decisions to understand how policy and service design may affect system-level and individual outcomes. We explore agent-based modeling, planning, and simulation of boundedly rational behavior.
Related projects: BRAMA and Ontology of Social Service Needs.
Applications: Communities, Services, and Public Health
We work with interdisciplinary collaborators on problems such as service needs, social impact measurement, public health intelligence, and access to urban data. These settings test whether computational methods are interpretable, useful, and appropriate for real-world decision-making.
Related projects: Community Justice Advocates, PHAC Substance Use Monitoring, Canadian Urban Data Catalogue, and Common Impact Data Standard.
Browse the full project portfolio or contact us about collaboration.