<![CDATA[The classification process is the process of labeling a class of data sets that do not have a class label yet. In the classification process there will always be uncertainty. The uncertainty here is that there is a possibility that the label chosen is not right and causing doubts. One method that can be used to overcome uncertainty is to use the Fuzzy and Dempster-Shafer (DS) methods. This research combines the fuzzification step to get the Belief value which will then be used in the DS classification calculation. This research aims to determine the effect of using different types of Membership Function in the classification process and the optimal parameters used in each MF. This research will combine the Fuzzification step from Fuzzy method to obtain the Belief value which will be then used in DS classification calculation. The Fuzzification step uses triangle and bell Membership Function (MF) to produce Belief value for the class label. The MF curve parameter tests are divided into two parts, the first part is where the parameters of the center point of the curve are within the range of the input data and the second part is where the parameters of the center point of the curve are outside the range of the input data. The result show that the optimal parameters for the triangle MF are a1 = a2 = 4, b1 = b2 = 10, c1 = 0 and c2 = 11 and parameters for the bell MF are a1 = -11, b1 = 0, c1 = 11 and a2 = 0, b2 = 1, c2 = 22. 2. The results of the research also show that the shape of bell MF with an accuracy of 88.87% is better than triangle MF.]]>
Copyrights © 2020
