Solar proton events prediction using learning vector quantity network

In order to improve the prediction accuracy, learning vector quantity (LVQ) was applied to construct solar proton event prediction model. LVQ is new type of neutral network based on competitive learning rule, which takes a supervised learning model. The structure of LVQ is a two layers neutral network. The input unit is the predictors which are some active region parameters correlated to proton event. The output unit is the class label of proton occurrence or not. Test result shows that the prediction model has high forecast accuracy and the LVQ is an effectual prediction method.