Optimize Takagi Sugeno Kang fuzzy system type 1 combination stochastic gradient descent with rough set.

Classification is a data analysis in which a model is being extracted in order to explain or differentiate the concept of each various classes of data. Fuzzy classification is one of the classification methods that is often being used because of its superiority non-flexible calculation so that it's possible to calculate uncertain possibility. In fuzzy method, there is a fuzzy set with the function to widen the range in characteristic function so that the function includes real number in [0,1] interval. One of recently well-known fuzzy method is Fuzzy Takagi Sugeno Kang (TSK). TSK method resulted in direct accurate output because of its rule where it use polynomial as consequences. In this research, the method used is fuzzy TSK with Rough Set which was optimized through Stocashatic Gradient Descent (SGD). The data used is secondary data obtained from database and being processed with Software Jupyter Notebook with Python as its programming language. The data is composed of cardiovascular data with 14 variables (13 input variables and 1 output variable). The model is evaluated by Root Mean Square Error (RMSE) and showed the average correspondence of the sum of squared prediction value and through observation of that data, resulting in 15,6. Based on the RMSE value obtained, it can be concluded that the used model is in a good category. [ABSTRACT FROM AUTHOR]