A progressive learning method for symbols recognition
Sabine Barrat, Salvatore Tabbone · 2007
This paper deals with a progressive learning method for symbols recognition which improves its own recognition rate when new symbols are recognized in graphics documents. We propose a discriminant analysis method which provides allocation rules from learning samples with known classes. However a discriminant analysis method is efficient only if learning samples and data are defined in the same conditions but it is rare in real life. In order to overcome this problem, a conditional vector is added to each observation to take into account the parasitic effects between the data and the learning samples. We propose also an adaptation to consider the user feedback.