One of the most significant steps in fuzzy modeling of a complex system is Structure Identification. Efficient structure identification requires good approximation of the effective input data. Misclassification of effective input data can significantly degrade the efficiency of the inference of the fuzzy model. In this paper we present a modification to the Sugeno-Yasukawa modeler [1] to improve structure identification by increasing the accuracy of effective input data detection. We improved Sugeno-Yasukawa Modeler by modifying the algorithm in two ways. Firstly, we used a new Trapezoid Approximation method based on [2] to improve estimation of membership functions. Secondly we change the modeling process of modeling. There exist some intermediate models in the Sugeno-Yasukawa modeling process, a combination of which will result in the final fuzzy model of the system. In the original modeling process, parameter identification is only done for the final fuzzy model. By doing the parameter identification for the intermediate fuzzy models, we have improved the accuracy of these intermediate models. The RC (Regularly Criterion) error has been reduced for intermediate fuzzy models and the MSE decreased without using the new Trapezoid Approximation method. By using the new trapezoid method, the RC value for the intermediate models and MSE for the final model improved even more. This accuracy increase, result in a better detection of effective input data among input data records of a system.