Multi-objective supervised clustering GA and microthermal climate forecast

摘要

A new multi-objective supervised clustering genetic algorithm is proposed in this paper. Training samples are supervised clustered by attribute similarity and class label. The number and center of class family can be determined automatically by using the fitness vector function. The two key elements have optimization nature and can be unaffected by subjective factors. Use the nearest neighbor rule and the class label to estimate the class families of test samples. The early warning model is implemented by C#, using the data of summery abnormal microthermal climate in Zhejiang province. The experiment results indicate that this algorithm has a unique intelligence and high accuracy.