In this paper, three different neural network structure which are Self Organizing Map (SOM), Probablistic Neural Network (PNN) and Radial Basis Function (RBF) were applied to the Escherichia coli benchmark and their efficiency in classifying the dataset has been obtained. Then the dataset is applied to the proposed coactive neuro-fuzzy inference system (CANFIS) model integrated with genetic algorithm and better classification with less MSE is obtained when tested using replicative testing.