This Article 
   
 Share 
   
 Bibliographic References 
   
 Add to: 
 
Digg
Furl
Spurl
Blink
Simpy
Google
Del.icio.us
Y!MyWeb
 
 Search 
   
Sequential Decision Models for Expert System Optimization
September-October 1997 (vol. 9 no. 5)
pp. 675-687

Abstract—Sequential decision models are an important element of expert system optimization when the cost or time to collect inputs is significant and inputs are not known until the system operates. Many expert systems in business, engineering, and medicine have benefited from sequential decision technology. In this survey, we unify the disparate literature on sequential decision models to improve comprehensibility and accessibility. We separate formulation of sequential decision models from solution techniques. For model formulation, we classify sequential decision models by objective (cost minimization versus value maximization) knowledge source (rules, data, belief network, etc.), and optimized form (decision tree, path, input order). A wide variety of sequential decision models are discussed in this taxonomy. For solution techniques, we demonstrate how search methods and heuristics are influenced by economic objective, knowledge source, and optimized form. We discuss open research problems to stimulate additional research and development.

[1] L. Breiman, J. Friedman, R. Olshen, and C. Stone, Classification and Regression Trees.Belmont, Calif.: Wadsworth, 1984.
[2] J. Breese and D. Heckerman, "Decision-Theoretic Case-Based Reasoning," IEEE Trans. Systems, Man, and Cybernetics, vol. 26, no. 6, pp. 838-842, Nov. 1996; also available as Microsoft Technical Report MSR-TR-95-03.
[3] J. Breese and D. Heckerman, "Decision Theoretic Troubleshooting: A Framework for Repair and Experiment," Proc. 12th Conf. Uncertainty in Artificial Intelligence, Aug. 1996; also available as Microsoft Technical Report MSR-TR-967-069.
[4] J. Breese and D. Heckerman, "Decision Theoretic Troubleshooting: A Framework for Repair and Experiment," Proc. 12th Conf. Uncertainty in Artificial Intelligence, Aug. 1996; also available as Microsoft Technical Report MSR-TR-96-06.
[5] A. Bagchi and A. Mahanti, "AND/OR Graphs Heuristic Search Methods," J. ACM, vol. 32, no. 1, pp. 28-51, 1985.
[6] V. Barker and D. O'Connor, "Expert Systems for Configuration at Digital: XCON and Beyond," Comm. ACM, Vol. 32, No. 3, Mar. 1989, pp. 298-318.
[7] A. Balakrishnan and A. Whinston, "Information Issues in Model Specification," Information Systems Research, vol. 2, no. 4, pp. 263-286, 1991.
[8] F. Coenen and T. Bench-Capon, Maintenance of Expert Systems, APIC Series, no. 40. San Diego, Calif.: Academic Press, 1993.
[9] J. Cendrowska, "PRISM: An Algorithm for Inducing Modular Rules," Int'l J. Man-Machine Studies, vol. 27, pp. 349-370, 1987.
[10] S. Chakrabarti, S. Ghose, and S. DeSarkar, "Admissibility of AO* When Heuristics Overestimate," Artificial Intelligence, vol. 34, NorthHolland, pp. 97-113, 1988.
[11] R. Davis, "Content Reference: Reasoning About Rules," Artificial Intelligence, no. 15, pp. 223-239, 1980.
[12] B. Dos Santos and V. Mookerjee, "Expert System Design: Minimizing Information Acquisition Costs," Decision Support Systems, vol. 9, NorthHolland, pp. 161-181, 1993.
[13] J. de Kleer and B. Williams, "Diagnosing Multiple Faults," Artificial Intelligence, vol. 32, pp. 97-130, Elsevier, 1987.
[14] K. Fu, "Artificial Intelligence," Handbook of Human Factors, G. Salvendy, ed., John Wiley&Sons, 1987.
[15] M. Garey, "Optimal Binary Identification Procedures," SIAM J. Applied Math., vol. 23, no. 2, pp. 173-186, 1981.
[16] M. Garey and R. Graham, "Performance Bounds on the Splitting Algorithm for Binary Testing," Acta Informatica, vol. 3, pp. 347-355, 1974.
[17] S. Ganapathy and V. Rajaraman, "Information Theory Applied to the Conversion of Decision Tables to Computer Programs," Comm. ACM, vol. 16, no. 9, pp. 532-539, Sept. 1973.
[18] E. Horvitz and M. Barry, "Display of Information for Time-Critical Decision Making," Proc. 11th Conf. Uncertainty in Artificial Intelligence, Montreal, Aug. 1995.
[19] D. Heckerman, J. Breese, and K. Rommelse, "Troubleshooting Under Uncertainty," Comm. ACM, Mar. 1995.
[20] D. Heckerman, E. Horvitz, and B. Middleton, "An Approximate, Nonmyopic Computation for Value of Information," Proc. Seventh Conf. Uncertainty in Artificial Intelligence, Univ. of California, Los Angeles, pp. 135-141, July 1992.
[21] D. Heckerman, E. Horvitz, and B. Nathwani, "Toward Normative Expert Systems: Part I. The PathFinder Project," Methods of Information in Medicine, vol. 31, pp. 90-115, 1992.
[22] D. Heckerman and B. Nathwani, "An Evaluation of the Diagnostic Accuracy of PathFinder," Computers and Biomedical Research, vol. 25, pp. 56-74, 1992.
[23] L. Hyafil and R. Rivest, "Constructing Optimal Binary Decision Trees is NP-Complete," Information Processing Letters, vol. 5, no. 1, pp. 15-17, May 1976.
[24] D. Huffman, "A Method for the Construction of Minimum Redundancy Codes," Proc. IRE, vol. 40, no. 10, pp. 1,098-1,101, Oct. 1952.
[25] K. Irani, J. Cheng, U. Fayyad, and Z. Qian, "Applying Machine Learning to Semiconductor Manufacturing," IEEE Expert, vol. 8, no. 1, pp. 41-47, Feb. 1993.
[26] M. James, Classification Algorithms. John Wiley&Sons, 1985.
[27] R. Johnson, "An Information Theory Approach to Diagnosis," Proc. Sixth Symp. Reliability Quality Control, pp. 102-109, 1960.
[28] V. Jacob, J. Moore, and A. Whinston, "Artificial Intelligence and the Management Science Practitioner: Rational Choice and Artificial Intelligence," Interfaces, vol. 18, no. 4, pp. 24-35, 1988.
[29] P. King and R. Johnson, "Some Comments on the Use of Ambiguous Decision Tables and Their Conversion to Computer Programs," Comm. ACM, vol. 16, no. 5, pp. 287-290, May 1973.
[30] J.L. Kolodner, “Improving Human Decision Making Through Case-Based Decision Aiding,” AI Magazine, Vol. 12, Sum. 1991, pp. 52-68.
[31] T. Liang, "A Composite Approach to Inducing Knowledge for Expert System Design," Management Science, vol. 38, no. 1, pp. 1-17, 1992.
[32] A. Bagchi and A. Mahanti, "AND/OR Graphs Heuristic Search Methods," J. ACM, vol. 32, no. 1, pp. 28-51, 1985.
[33] V. Mookerjee and B. Dos Santos, "Inductive Expert System Design: Maximizing System Value," Information Systems Research, vol. 4, no. 2, pp. 111-140, June 1993.
[34] J. Mingers, “An Empirical Comparison of Pruning Methods for Decision Tree Induction,” Machine Learning, vol. 4, no. 2, pp. 227-243, 1989.
[35] A. Martelli and U. Montanari, "Additive And/Or Graphs," Proc. Third Int'l Conf. Artificial Intelligence,Stanford, Calif., pp. 1-11, Aug. 1973.
[36] A. Martelli and U. Montanari, "Optimizing Decision Trees Through Heuristically Guided Search," Comm. ACM, vol. 21, no. 12, pp. 1,025-1,039, Dec. 1978.
[37] V. Mookerjee and M. Mannino, "Redesigning Case Retrieval Systems to Reduce Information Acquisition Costs," Information Systems Research, vol. 8, no. 1, pp. 51-68, Mar. 1997; URL:http://www.isi.edu/gost/publications/kerberos-neuman-tso.htmlhttp:/ /csdl.computer.org/ comp/trans/tk/1993/04/k0564abs.htmhttp:/ /www.cudenver.edu~mmannino.
[38] M. Mannino and V. Mookerjee, "Redesigning Expert Systems: Heuristics for Efficiently Generating Low Cost Information Acquisition Strategies," INFORMS J. Computing, under review; URL:http://www.cudenver.edu~mmannino.
[39] V. Mookerjee and M. Mannino, "Mean-Variance Tradeoffs in Inductive Expert System Construction," Information Systems Research, under review; URL:http://www.cudenver.edu~mmannino.
[40] V. Mookerjee, M. Mannino, and B. Gilson, "Improving the Performance Stability of Inductive Expert Systems Under Input Noise," Information Systems Research, vol. 6, no. 4, Dec. 1995.
[41] M. Mannino, V. Mookerjee, and B. Gilson, "Overcoming Non-Monotonicity in Bayesian Belief Networks," IEEE Trans. Knowledge and Data Engineering, under review; URL:http://www.cudenver.edu~mmannino.
[42] V. Mookerjee, "Debiasing Training Data for Inductive Expert System Construction," INFORMS J. Computing, under review.
[43] B.M.E. Moret,“Decision trees and diagrams,” Computing Surveys, vol. 14, no. 4, pp. 593-623, 1982.
[44] J. Moore, W. Richmand, and A. Whinston, "A Decision Theoretic Approach to Information Retrieval," ACM Trans. Database Systems, vol. 15, no. 3, pp. 311-340, 1990.
[45] J. Moore, H. Rao, and A. Whinston, "Multi-Agent Resource Allocation: An Incomplete Information Perspective," IEEE Trans. Systems, Man, and Cybernetics, vol. 24, no. 8, Aug. 1994.
[46] J. Moore and A. Whinston, "A Model of Decision Making with Sequential Information Acquisition—Part I," Decision Support Systems, vol. 2, no. 4, NorthHolland, pp. 285-307, 1986.
[47] J. Moore and A. Whinston, "A Model of Decision Making with Sequential Information Acquisition—Part II," Decision Support Systems, vol. 3, no. 1, NorthHolland, pp. 47-72, 1987.
[48] H. Myers, "Compiling Optimized Code from Decision Tables," IBM J. Research and Development, vol. 16, no. 5, pp. 489-503, Sept. 1972.
[49] H. Newquist, "No Summer Returns," AI Expert, Oct. 1990.
[50] M. Nunez, "The Use of Background Knowledge in Decision Tree Induction," Machine Learning, vol. 6, pp. 231-250, 1991.
[51] K. Patttipati and M. Alexandridis, "Application of Heuristic Search and Information Theory to Sequential Fault Diagnosis," IEEE Trans. Systems, Man, and Cybernetics, vol. 20, no. 4, pp. 872-887, July/Aug. 1990.
[52] J. Pearl, Heuristics: Intelligent Search Strategies for Computer Problem Solving, Addison-Wesley, Reading, Mass., 1984.
[53] S. Pollack, H. Hicks, and W. Harrison, Decision Tables: Theory and Practice.New York: Wiley, 1971.
[54] J.R. Quinlan,"Induction of decision trees," Machine Learning, vol. 1, pp. 81-106, 1986.
[55] J.R. Quinlan,“Simplifying decision trees,” Int’l J. Man-Machine Studies, vol. 27, pp. 221-234, 1987.
[56] L. Reinwald and R. Soland, "Conversion of Limited-Entry Decision Tables to Optimal Computer Programs," J. ACM, vol. 13, no. 3, pp. 339-358, July 1966.
[57] K. Schwayder, "Extending the Information Theory Approach to Converting Limited-Entry Decision Tables to Computer Programs," Comm. ACM, vol. 17, no. 9, pp. 532-537, Sept. 1974.
[58] A. Saharia and G. Diehr, "A Refresh Scheme for Remote Snapshots," Information Systems Research, vol. 1, no. 3, 1990.
[59] R. Shachter, "Evaluating Influence Diagrams," Operations Research, vol. 34, no. 6, pp. 871-882, Nov-Dec. 1986.
[60] K. Schwayder, A. Kenney, and R. Ainslie, "Decision Tables: A Tool for Tax Practitioners," The Tax Advisor, vol. 3, no. 6, pp. 336-345, June 1972.
[61] E. Simoudis and J. Miller, "The Application of CBR to Help Desk Applications," Proc. Workshop Case-Based Reasoning, pp. 25-36, 1991.
[62] H. Schumacher and K. Sevcik, "The Synthetic Approach to Decision Table Conversion," Comm. ACM, vol. 19, no. 6, pp. 343-351, June 1976.
[63] R. Smith and C. Scott, Innovative Applications of Artificial Intelligence, vol. 3. Menlo Park, Calif.: AAAI Press, 1991.
[64] C. Shannon and C. Weaver, The Mathematical Theory of Comm., Univ. of Illinois Press, 1949 (published, 1964).
[65] R. Tiong and T. Koo, "Selecting Construction Formwork: An Expert System Adds Economy," Expert Systems, Spring 1991.
[66] M. Verhelst, "The Conversion of Limited Entry Decision Tables to Optimal and Near Optimal Flowcharts: Two New Algorithms," Comm. ACM, vol. 15, no. 11, pp. 974-980, Nov. 1972.
[67] P. Varshney, P. Hartman, and J. DeFaria, "Application of Information Theory to Sequential Fault Diagnosis," IEEE Trans. Computers, vol. 31, no. 2, pp. 164-170, 1982.
[68] J. Jarvis, "An Information Theory Approach to Diagnosis," Proc. Sixth Symp. Reliability Quality Control, pp. 102-109, 1960.
[69] P. Turney, "Cost Sensitive Classification: Empirical Evaluation of a Hybrid Genetic Decision Tree Induction Algorithm," J. Artificial Intelligence Research, vol. 2, pp. 369-409, 1995.

Index Terms:
Sequential decision models, expert system optimization, information acquisition costs, decision costs and benefits, cost minimization, value maximization, decision tables and trees.
Citation:
Vijay S. Mookerjee, Michael V. Mannino, "Sequential Decision Models for Expert System Optimization," IEEE Transactions on Knowledge and Data Engineering, vol. 9, no. 5, pp. 675-687, Sept.-Oct. 1997, doi:10.1109/69.634747
Usage of this product signifies your acceptance of the Terms of Use.