Basser Seminar Series

Context Modeling: Representing Practices instead of Procedures

Professor Patrick Brezillon
Laboratoire d'Informatique de Paris 6

Visiting speaker series supported by EII Taskforce on Context-Aware Computing

Tuesday 30 October 2007, 4-5 pm, **Note different day

School of IT Building, Lecture Theatre Room 123, Level 1


Contextual Graphs are a context-based formalism used in various real-world applications. They allow a uniform representation of elements of reasoning and of contexts for describing different human tasks such as troubleshooting and interpretation. A contextual graph represents a task realization. Its paths represent the different ways of reaching this realization, each way corresponding to a practice developed by an actor realizing the task. The seminar will revisit the classical distinction between prescribed and effective tasks, procedures versus practices, logic of functioning versus logic of use, etc. in the light of this formalism. The position of the practice model with respect to the task model will be discussed and experimented across several examples including the troubleshooting a problem with a DVD player.

Speaker's biography

The research of Dr P. Brezillon focuses since 1992 on the notion of context, and he now is considered as one of the main leader in the community interested by context by participating in the organization of the series of international and interdisciplinary conference on context (CONTEXT), giving invited talks, initiating the French Association for Context and beginning to develop a portal of Context After the design and development of a coherent conceptual framework for context modeling, his ideas are now formalized in a context-based formalism-called Contextual Graphs-for representing in a uniform way elements of reasoning and of contexts ( This formalism is used in 15 different domains concerned by reasoning in context across the world. His research is taught in a Specialization (Management of Knowledge, Contents and Contexts). He published about 300 papers alone or with a grand total of 120 co-authors in international conference and international journals (e.g., IEEE Expert, AI Magazine, The Knowledge Engineering Review, the International Journal on Human-Computer Studies, etc.). Information is available at