A Multiclass Classification Method Based on Multiple Pairwise Classifiers

In this paper, a new method of composing a multiclass classifier using pairwise classifiers is proposed. A “Resemblance Model” is exploited to calculate a posteriori probability for combining pairwise classifiers. We proved the validity of this model by using approximation of a posteriori probability formula. Using this theory, we can obtain the optimal decision. An experimental result of handwritten numeral recognition is presented, supporting the effectiveness of our method.

• Type: Conference
• Year: 2003
• Where: ICDAR
• Authors: Tomoyuki Hamamura, Hiroyuki Mizutani, Bunpei Irie
• Keywords: Document Analysis, ICDAR 2003, Pairwise Classifiers, Posteriori Probability, Posteriori Probability Formula