Information Retrieval from Historical Corpora

With the increasing number of documents that are available in digital form, also the number of digital historical documents is increasing (Berkvens, 2001). It cannot be assumed that standard IR systems perform well on historical documents: historical texts differ from modern texts in three ways (Hüning, 1996; Van Der Horst and Marschall, 1989): (a) vocabularies have changed, (b) spelling has changed (sometimes to the extent that two variants of the same word are not even recognizable as such), and (c) spelling used to be highly inconsistent (in the Netherlands until the 19th century). The goals of this research were to identify the bottlenecks of information retrieval from historical corpora and to find solutions for these bottlenecks. Section 1 describes our efforts to further identify the bottlenecks, section 2 examines potential solutions, and section 3 sketches our approach to alleviate the bottlenecks. Finally section 4 draws conclusions and provides directions for future research.