Innovative Criteria for Input Selection
Madasu Hanmandlu, M.F. Azeem, Vamsi Krishna Madasu · 2003
In this paper an attempt is made to derive new criteria for input selection of dynamic systems using the fuzzy curve approach. The Approximate Fuzzy Data Model (AFDM), the output of which is the fuzzy curve, is shown to be a special case of the Generalized Fuzzy Model (GFM). Moreover, AFDM is proved to be an unconditional expectation of the output thus linking fuzzy rules with probability. The validity of the criteria for input selection has been studied on GFM by means of significance of inputs, which is determined from the ratio of change in the output of AFDM to the range of the actual output. The complexity of the criteria has been proved to be of the order of O(n), which is a significant achievement in comparison to the complexity of the existing criteria.