Predicting System for the Behavior of Consumer Buying Personal Car Decision by Using SMO

Current consumer behavior traders are used tocreate decision-making tools for entrepreneurs to produce products or services that meet consumer needs. For this reason, therefore focusing on data analysis using SMO techniques for predict system for the behavior of consumer buying personal car decision based on a total of 1,110 data obtained, with a total of 6 relevant car trading information features, consisting of income customers, type of car, down payment/ cash booking, decision results requires that the answer in the forecast is divided into two classes, namely the class of buying cars and not buying cars When dividing the data into 50% training set and using 50% test set, 555 training set will be used and 555 test sets can be used to enter the learning and testing process. By random 50/50 in the selection of algorithms to be tested and the accuracy rate is 95.13%