bokstaffua, bokstaffwa, bokstafwa, bokstaua, bokstawa... Towards lexical link-up for a corpus of Old Swedish
Yvonne Adesam, Malin Ahlberg, Gerlof Bouma · 2012
We present our ongoing work on handling spelling variations in Old Swedish texts, which lack a standardized orthography. Words in the texts are matched to lexica by edit distance. We compare manually compiled substitution rules with rules automatically derived from spelling variants in a lexicon. A pilot evaluation showed that the second approach gives more correct matches, but also more false positives. We discuss several possible improvements. The work presented is a step towards natural language processing of Old Swedish texts.