Dr. Ghassan M. Azar, M.S.W., Ph.D.
Computer Science College Professor and Program Director
Faculty Senate Chair, since Fall 2003
Office hours by appointment
Dr. Azar has been a member of the Mathematics and Computer Science since fall 2000. He has over twenty-five years of practical experience in the computing field. He established Computer Software, Inc. in 1992 managing up to 15 consultants. CSI grossed over a million dollar a year providing services for BellSouth, Electronic Data Systems, Ford Motor Company, IBM, Pacific Bell, Southern Bell, Visteon, American Express and an international union.
Dr. Azar earned from Wayne State University a B.A 1982, M.S. 1986, and Ph.D. 1992 in Computer Science Parallelizing Compilers. His research interests are Artificial Intelligence, Parallel Processing, Data Warehousing and Data Mining.
He recently earned from the University of Michigan in Ann Arbor a Master in Social Work MSW specializing in Mental Health and Individual Therapy.
Dr. Azar’s research interests are:
1) Develop a semi-automated system to help users in classifying mental health illness.
In societies, where there is abundance of qualified and educated classifiers in medical diagnosis, biases lead to inaccurate classification in general and specifically for mental health illnesses . Mentally ill individuals are judged by the way they look and act without deep understanding of their behaviors and symptoms . This phenomenon occurs for two reasons. First, the pressure applied by providers and insurance companies to diagnose individuals within the first 30 minutes of their initial visit to obtain and lock insurance coverage. Hence, insurance coverage may sway a classifier’s decision to secure specific number of treatment sessions for a patient. The second is the lack of knowledge that therapists and/or psychiatrists possess to weed through the many mental health illnesses identified in . Improper diagnosis leads to prescribing and dispensing wrong medications and delivering wrong therapeutic solutions to a client. This improper diagnosis, at worst cases, may lead to irreversible deterioration in the client’s mental health status including hospitalization and/or pre-mature death.
We have identified a specific problem in diagnosing and classifying individuals in the mental health field . In less fortunate countries where there is a shortage of qualified and educated classifiers, the classification problem or the lack thereof, is multiplied several folds. The issue at hand is having unqualified individuals performing diagnosis and classification of mentally ill individuals.
 J. Shedler, “The illusion of Mental Health,” American Psychologist, Vol 48 No 11 1117-1131, 1993.
 S. Fernando, “Mental Health in a Multi-Ethnic Society,” A Multi-Disciplinary Handbook, Routledge
 American Psychiatric Association. [DSM-IV-TR], (2000). Diagnostic and statistical manual of mental disorders (Revised 4th ed.). Washington, DC: Author
2) Brain simulation:
Like the fictional brain in a vat, artificial minds are useless if they are not connected to the real world. Embodied cognitive systems require bodies and worlds where they can interact, explore, and learn. An open framework which connects minds to worlds is therefore of paramount importance. This undertaking involves both building the framework, and building an initial cognitive system which can be instantiated on demand. The Piagetian Autonomous Modeler is a cognitive system architecture which constructs an internal model of its world in the form of a database of neural propositions. The architecture is compatible with Global Workspace Theory and Society of Mind Theory, and is primarily inspired by:
Piaget J. and Roslin, A. 1978. The Development of Thought: The Equilibration of Cognitive Structures. Viking Press.
Piaget, J.; Brown, T.; and Thampy K.J. 1985. Equilibration of Cognitive Structures: The Central Problem of Intellectual Development.
Drescher, G. 1991. Made Up Minds: A Constructivist Approach to Artificial Intelligence. MIT Press.
Indurkhya, B. 1992. Metaphor and Cognition: An Interactionist Approach. Kluwer Academic.
Pickett, M.; Miner, D., Oates, T. 2005. Essential Phenomena of General Intelligence.
Pickett, M.; Oates, T. 2005. The Cruncher: Automatic Concept Formation Using Minimum Description Length. Abstraction, Reformulation and Approximation, 6thInternational Symposium, SARA.
Morrison, C.T.; Oates, T.; King, G.W. 2001. Grounding the Unobservable in the Observable: The Role and Representation of Hidden State in Concept Formation and Refinement.
Heib, M.R. and Michalski, R.S. 1993. Multitype Inference in Multistrategy Task Adaptive Learning: Dynamic Interlaced Hierarchies. In Proceedings of the Second International Workshop on Multistrategy Learning, Harper’s Ferry, WV.
Alkharouf, N. and Michalski, R.S. 1996. Multi-Strategy Task Adaptive Learning Using Dynamically Interlaced Hierarchies in Proceedings of the Third International Conference on Multistrategy Learning.
Hausser, R. 2010. A Computational Model of Natural Language Communication: Interpretation, Inference and Production in Database Semantics. Springer Verlag. Berlin Heidelberg.
Miller, M. S. P. 2011. Piagetian Autonomous Modeller. InProceedings of the AISB 2011 Symposium on Computational Models of Cognitive Development: 32-39.
Miller, M. S. P. 2012a. Patterns for Cognitive Systems. In Proceedings of the Sixth International Conference on Complex, Intelligent and Software Intensive Systems (CISIS). Palermo Italy
Miller, M. S. P. 2012b. Serving Up Minds. (Unpublished.)
Miller, M. S. P. 2013. The Neural Proposition: Structures for Cognitive Systems. AAAI Press - 2013 AAAI Spring Symposium on Creativity and (Early) Cognitive Development - Stanford, CA, USA
Hammond, K. J. 1989. Case-Based Planning: Viewing Planning as a Memory Task. Academic Press Inc.: 14-29.
Lenat, D. B. and Guha, R. V. 1990. Building Large Knowledge Based Systems: Representation and Inference in the Cyc Project.