Finding canonical forms for historical German text

Historical text presents numerous challenges for contemporary natural language processing techniques. In particular, the absence of consistent orthographic conventions in historical text presents difficulties for any technique or system requiring reference to a fixed lexicon accessed by orthographic form. This paper presents two methods for mapping unknown historical text types to one or more synchronically active canonical types: conflation by phonetic form, and conflation by lemma instantiation heuristics. Implementation details and evaluation of both methods are provided for a corpus of historical German verse quotation evidence from the digital edition of the Deutsches Wörterbuch.