Non linear Prediction of Antitubercular Activity Of Oxazolines and Oxazoles derivatives Making Use of Compact TS-Fuzzy models Through Clustering with orthogonal least sqaure technique and Fuzzy identification system

The prediction of uncertain and predictive nonlinear systems is an important and challenging problem. Fuzzy logic models are often a good choice to describe such systems however in many cases these become complex soon. commonlly, too less effort is put into descriptor selection and in the creation of suitable local rules. Moreover, in common no model reduction is applied, while this may analyze the model by removing redundant data. This paper suggests a combined method that deal with these issues in order to create compact Takagi Sugeno (TS) models that can be effectively used to represent complex predictive systems. A new fuzzy clustering method is come up with for the identification of compact TS-fuzzy models. The best relevant consequent variables of the TS model are choosen by an orthogonal least squares technique based on the obtained clusters.For the selection of the relevant antecedent (scheduling) variables a new method has been developed based on Fisher's interclass separability basis. This complete approach is demonstrated by means of the Oxazolines and Oxazoles derivatives as antituberculosis agent for nonlinear regression benchmark. The results are compared with results obtained by neuro-fuzzy i.e. ANFIS algorithm and advanced fuzzyy clustering techniques i.e FMID toolbox .