{"title":"On the Parameter Optimization of Fuzzy Inference Systems","authors":"Erika Martinez Ramirez, Rene V. Mayorga","volume":18,"journal":"International Journal of Computer and Information Engineering","pagesStart":2024,"pagesEnd":2038,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/15939","abstract":"Nowadays, more engineering systems are using some\r\nkind of Artificial Intelligence (AI) for the development of their\r\nprocesses. Some well-known AI techniques include artificial neural\r\nnets, fuzzy inference systems, and neuro-fuzzy inference systems\r\namong others. Furthermore, many decision-making applications base\r\ntheir intelligent processes on Fuzzy Logic; due to the Fuzzy\r\nInference Systems (FIS) capability to deal with problems that are\r\nbased on user knowledge and experience. Also, knowing that users\r\nhave a wide variety of distinctiveness, and generally, provide\r\nuncertain data, this information can be used and properly processed\r\nby a FIS. To properly consider uncertainty and inexact system input\r\nvalues, FIS normally use Membership Functions (MF) that represent\r\na degree of user satisfaction on certain conditions and\/or constraints.\r\nIn order to define the parameters of the MFs, the knowledge from\r\nexperts in the field is very important. This knowledge defines the MF\r\nshape to process the user inputs and through fuzzy reasoning and\r\ninference mechanisms, the FIS can provide an \u201cappropriate\" output.\r\nHowever an important issue immediately arises: How can it be\r\nassured that the obtained output is the optimum solution? How can it\r\nbe guaranteed that each MF has an optimum shape? A viable solution\r\nto these questions is through the MFs parameter optimization. In this\r\nPaper a novel parameter optimization process is presented. The\r\nprocess for FIS parameter optimization consists of the five simple\r\nsteps that can be easily realized off-line. Here the proposed process\r\nof FIS parameter optimization it is demonstrated by its\r\nimplementation on an Intelligent Interface section dealing with the\r\non-line customization \/ personalization of internet portals applied to\r\nE-commerce.","references":"[1] Jang J.-S. R., Sun C.-T. Mizutani E. Neuro-Fuzzy and Soft Computing:\r\nA computational approach to learning and machine intelligence. Matlab\r\nCurriculum Series. Edit. Prentice Hall. 1997.\r\n[2] Martinez E. Mayorga R. V., \"An Architecture for the Coupling of\r\nIntelligent Computer Interfaces with Intelligent Systems: An Online\r\nInternet Portals Customization Application. Proceedings 4th\r\nANIROB\/IEEE-RAS Intl. Symposium on Robotics and Automation,\r\nQueretaro, Mexico, August, 25-27, 2004\r\n[3] Mayorga R.V. \"Towards Computational Sapience (Wisdom) and\r\nMetabotics: Intelligent \/ Sapient (Wise) Decision \/ Control, Systems, and\r\nMetaBots\", Proceedings 4th ANIROB\/IEEE-RAS Intl. Symposium on\r\nRobotics and Automation, Queretaro, Mexico, August, 25-27, 2004\r\n[4] Mayorga R.V. \"Towards Computational Sapience (Wisdom): A\r\nParadigm for Sapient (Wise) Systems\", Proceedings International\r\nConference on Knowledge Intensive Multi-Agent Systems, KIMAS-03,\r\nCambridge, MA, USA, September 30 - October 4, 2003.\r\n[5] Mayorga R. V. \"A Metabotics Paradigm for the Wise Design and\r\nOperation of a Human-Computer Interface, Proceedings 2nd\r\nANIROB\/IEEE-RAS Intl. Symposium on Robotics and Automation,\r\nMonterrey, Mexico, November 10-12, 2000.\r\n[6] Stewart, J., Calculus, Second Edition. Brooks\/Cole Publishing\r\nCompany. Pacific Grove, California.1991.\r\n[7] Matlab Help Tutorials on Fuzzy, Optimization and Symbolic Math\r\nToolboxes.\r\n[8] Yon J-h, Yang S-m, Jeon H-T. \"Structure Optimization of Fuzzy-Neural\r\nNetwork Using Rough Set Theory\" in 1999 IEEE International Fuzzy\r\nSystems Conference Proceedings. Korea, 1999.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 18, 2008"}