On-line cursive handwriting characterization using TF-IDF scores of graphemes

In this paper, we present an approach for characterizing the on-line cursive handwriting of different writers, which may consist in identifying the writer or his handwriting style. This method is inspired from information retrieval methods and is designed to be embedded in an adaptive word recognizer. We perform experiments assessing the effectiveness of the proposed method for writer identification. Additional preliminary experiments also show that the handwriting style can be used to personalize our cursive word recognizer, enabling the word recognition rates to be increased significantly even with a basic adaptive scheme, which is very encouraging. Keywords-Writer identification; handwriting style classification; adaptive cursive word recognition; on-line features; classification using TF-IDF.