Analysis of Different Writing Styles with the Self-Organizin Map

This work shows how a Self-Organizing Map (SOM) can be applied in the analysis of di erent handwriting styles. Handwriting styles are represented with vectors whose components re ect the tendencies of the writers to use certain prototypical styles for isolated alphanumeric characters. The study shows that the correlations between di erent writing styles congruent with prior human knowledge can be found with SOM. It turns out that the SOM can make a distinction between writers with cursive style or a mixture of print and block styles. The former group of subjects forms a clear cluster in a writing-style-space, and in their case, the correlations between the writing styles are very strong and understandable.