Early modern books written in Latin contain many abbreviations of common words that are derived from earlier manuscript practice. While these abbreviations are usually easily deciphered by a reader well-versed in Latin, they pose technical problems for full text digitization: they are difficult to OCR or have typed and — if they are not expanded correctly — they limit the effectiveness of information retrieval and reading support tools in the digital library. In this paper, I will describe a method for the automatic expansion and disambiguation of these abbreviations. Categories and Subject Descriptors Content, Collection Development, Document Processing General Terms Algorithms, Experimentation, Languages Keywords Digitization, Tagging Early Modern Texts, History of Science
Jeffrey A. Rydberg-Cox