@conference {13527, title = {Content Features for Logical Document Labeling}, booktitle = {Proc. SPIE Conference on Document Recognition and Retrieval X}, year = {2003}, month = {2003///}, pages = {189 - 196}, abstract = {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. }, author = {Liang,Jian and David Doermann} }