Automated Processing of Digitized Historical Newspapers: Identification of Segments and Genres

Abstract

Many historical newspapers are being digitized. We aim to support access to them via text analysis of the OCRd content. However, the OCR includes many errors; so extracting meaningful content from it is difficult. A pipeline of processing steps is proposed. Here, we describe the first two steps: segmentation and genre identification. The segmentation procedure based on headings was quite successful. Genre identification worked well for easily defined genre categories such as weather reports. We also propose additional techniques which may improve the accuracy still farther.

Publishing Historical Texts on the Semantic Web - A Case Study

“…A detailed report of the successes and failures of the task per class is presented. A pipeline for the automatic processing of historical newspapers is specified in [21]. The pipeline begins with OCR'd text in XML format, and includes article segmentation, genre and subject recognition, and finally event extraction.…”

“…The pipeline begins with OCR'd text in XML format, and includes article segmentation, genre and subject recognition, and finally event extraction. In [21] the results are reported for the first two steps, namely article segmentation and genre recognition. Their future plans include topic and event categorization from the material.…”

“…We will also try to extend our automatic semantic annotation tools to recognizing and annotating historical events that recur in the material and should be of special interest to end-users of news materials. In [21] a plan for categorization of events in historical newspaper archives is made. In our own work, we are creating an ontology of historical events [24] whose first small version is already in use in CultureSampo.…”