Abstract: Meteorological data mining has been a hot research spot recently. Combined classifiers can be used in collaborative computing to improve the efficiency and accuracy. Based on C4.5 classic algorithm and the bagging integrated method, it constructed a cooperative decision tree, and proposed a decision tree model for multi-classifiers to predict the air temperature. Experimental results show that the cooperative model for the prediction of local area atmospheric temperature, based on multi-classifiers of the decision tree, has higher accuracy than others.

Key words: Bagging; C4.5 algorithm; multi-classifiers; cooperative; air temperature prediction