A phonetic approach to handling spelling variations in medieval documents

  • Authors:

  • Mushtaq Ahmad;Nazim Rahman;Stefan Gruner

  • Affiliations:

  • University of Pretoria, South Africa;TEKRI, Athabasca University, Canada;University of Pretoria, South Africa

  • Venue:
  • Proceedings of the South African Institute of Computer Scientists and Information Technologists Conference on Knowledge, Innovation and Leadership in a Diverse, Multidisciplinary Environment
  • Year:
  • 2011

Quantified Score

Hi-index 0.00

Visualization

Abstract

Spelling variations pose a critical obstacle in the study, comprehension, and translation of medieval manuscripts. In this short paper we describe a new process and tool, POM (Phonetic Orthography Mapper), we developed to map spelling variations to standard German orthography used today. The tool is based on phonetic analysis and machine learning techniques. POM was applied to more than 20,000 digitalized German medieval manuscripts and we were able to correctly map more than 60,000 spelling variations. The research described in this short paper is part of a larger interdisciplinary project in the "digital humanities", particularly about software-tool support for the handling of historic central-European documents written in medieval German dialects.