Toward Opitmal Fuzzy Associative Systems Using Interactive Self-Organizing Maps and Multi-layer Feed Forward Principles
The objective is to find optimal fuzzy membership functions for any given data without using information external to the data set itself. This paper only reports research using a multi-layer feed forward architecture with one input variable. Target variables are the cumulative distribution of the input data and the constant one. Although learning is supervised, values are derived only from input data itself and the state of the network.
T. E. Sandidge and C. H. Dagli, "Toward Opitmal Fuzzy Associative Systems Using Interactive Self-Organizing Maps and Multi-layer Feed Forward Principles," ANNIE 1998, American Society of Mechanical Engineers (ASME), Jan 1998.
Engineering Management and Systems Engineering
Keywords and Phrases
Fuzzy Membership Function; Input Data; One Input Variable
Article - Conference proceedings
© 1998 American Society of Mechanical Engineers (ASME), All rights reserved.