SIE 474 - Decision Support Systems

SIE 474 - Decision Support Systems (3) Building, testing and evaluating expert systems, computer systems that emulate the human and draw conclusions based on incomplete or inaccurate data. Each student will build a decision support system using commercially available expert systems shells. Graduate-level requirements include a strong testing and validation study of the student’s expert system.

Ignizio, J.P. (1991) Introduction to Expert Systems: the development and implementation of rule-based expert systems. McGraw Hill.

Linder, Jane: Expert systems, A Note. Harvard Business School Press, 1989.

Prerau, David S.: Selection of an appropriate domain for an expert system. AI Magazine, summer 1985.

Kowalski, Allan et al: Pitch Expert: A problem solving system for Kraft mills. AI Magazine, Fall 1993.

Rumelhart D., Widrow, B., Lehr, M: The basic idea in neural networks. CACM, March 1994.

Bahill, A.T., Bharathan, K., Curlee, R.R.: How the testing techniques for a decision support system changed over nine years. IEEE Transactions on SMC, December, 1995.

Kalyanraman Bharathan

1. Student should have some experience in working with computers
2. Minimal keyboard skills
3. Enthusiasm for new and interesting problem areas and disciplines

The prerequisites are sufficiently general to not require verification.

To provide the foundation for good, elegant, and realistic knowledge engineering among students; providing a perspective on the areas where decision support tools are an asset and where they are not; providing the tools needed to assess the feasibility for the design and implementation of decision support systems, with special reference to expert systems.

1. What is a decision support system?
2. What is an expert system?
3. What is the environment they are good for - What problems are well suited for them?
4. How is knowledge extracted?
5. How is knowledge represented?
6. How do we deal with uncertainty?
7. How do we deal with partial information?

1. Identifying An Expert
2. Knowledge Extraction Techniques
3. Designing A Knowledge Base In M.4
4. Inference Methods: Backward Chaining, Forward Chaining, Neural Nets
5. Simple Decision Support Systems
6. Data Exploration, Data Warehouses And Data Mining
7. Learning The M.4 Expert System Shell
8. Dealing With Uncertainty

Class Requirements:

1. Attendance
2. Homework assignments approximately every 2 weeks
3. 2 midterm examinations and a final
4. Project presentation and submission

The project is built using M.4

Laboratory Projects: None

1. Homework evaluation
2. Examination
3. Project presentation and submission

Contribution to program objectives: Goals 2, 3, 4, 5

Prepared by: Kalyanraman Bharathan   Date: April 14, 1998