A labeling approach for automatic recognition of Tables of Contents (ToC) is described in this paper. A prototype is used for electronic consulting of scientific papers in a digital library system named Calliope. This method operates on a roughly structured ASCII file, produced by OCR. The recognition approach operates by text labeling without using any a priori model. Labeling is based on a Part of Speech Tagging (PoS) which is initiated by a primary labeling of text component using some specific dictionaries.