A prototype classification method and its use in a hybrid solution for multiclass pattern recognition

In order to combine a fast multiclass classification method with an effective binary classification method, we have developed a prototype learning/matching scheme that can be integrated with support vector machines (SVM) for vector-matching applications. This prototype classification method employs a learning process to determine both the number and the location of prototypes. The learning process decides whether to stop adding prototypes according to a certain termination condition, and also adjusts the location of prototypes using either the K-means (KM) or the fuzzy c-means (FCM) clustering algorithms. When the prototype classification method is applied, the SVM method can be used to post-process the top-rank candidates obtained during the prototype learning or matching process. We apply this hybrid solution to handwriting recognition. Our experi-