Searching for historical word-forms in a database of 17th-century English text using spelling-correction methods

摘要

This paper discusses the application of algorithmic spelling-correction techniques to the identification of those words in a database of 17th century English text that are most similar to a query word in modern English. The experiments have used n-gram matching, non-phonetic coding and dynamic programming methods for spelling correction, and have demonstrated that high-recall searches can be carried out, although some of the searches are very demanding of computational resources. The methods are, in principle, applicable to historical texts in many languages and from many diffeent periods.