Fuzzy support vector machine based on improved sequential minimal optimization algorithm

Sequential minimal optimization (SMO) is quite an efficient algorithm for training the support vector machine, which can resolve the optimization question of a fuzzy support vector machine (FSVM), but it still requires a large amount of computation time for solving large size problems. This paper proposes an improved SMO using for fast training a fuzzy support vector machine. The training is realized by utilizing the inner loop and the outer loop until all examples obey KKT conditions. Using this method for modeling an actual penicillin fermentation process, the result shows that this method can not only make shorten training time, but also have a better predictive precision; on the other hand, this method can meet the request of process online survey. By analyzing the simulating result of the penicillin concentration, this method can advance the precision of simulating result affectively when the process parameters are changing greatly.