TY - CONF T1 - Content Features for Logical Document Labeling T2 - Proc. SPIE Conference on Document Recognition and Retrieval X Y1 - 2003 A1 - Liang,Jian A1 - David Doermann AB - The use of content features extracted from recognized text is valuable in labeling logical elements in documents withoutrigid layout structure, like business letters. This paper discusses a model-based approach to combining content features with other geometrical and presentation features for logical labeling. Models are automatically initialized and adaptively improved using training samples. Satisfactory experimental results are presented. JA - Proc. SPIE Conference on Document Recognition and Retrieval X ER -