Error-tolerant Finite State Recognition with Applications to Morphological Analysis and Spelling Correction

TL;DR

This paper introduces error-tolerant finite state recognition algorithms for morphological analysis and spelling correction, demonstrating their efficiency and applicability across multiple languages, including Turkish, English, and European languages.

Contribution

It presents novel algorithms for error-tolerant recognition using finite state transducers, enabling accurate morphological analysis and spelling correction across diverse languages.

Findings

Efficient candidate generation within 10-45 ms for large word lists

Successful application to Turkish agglutinative morphology

Effective spelling correction for multiple European languages

Abstract

Error-tolerant recognition enables the recognition of strings that deviate mildly from any string in the regular set recognized by the underlying finite state recognizer. Such recognition has applications in error-tolerant morphological processing, spelling correction, and approximate string matching in information retrieval. After a description of the concepts and algorithms involved, we give examples from two applications: In the context of morphological analysis, error-tolerant recognition allows misspelled input word forms to be corrected, and morphologically analyzed concurrently. We present an application of this to error-tolerant analysis of agglutinative morphology of Turkish words. The algorithm can be applied to morphological analysis of any language whose morphology is fully captured by a single (and possibly very large) finite state transducer, regardless of the word formation processes and morphographemic phenomena involved. In the context of spelling correction, error-tolerant recognition can be used to enumerate correct candidate forms from a given misspelled string within a certain edit distance. Again, it can be applied to any language with a word list comprising all inflected forms, or whose morphology is fully described by a finite state transducer. We present experimental results for spelling correction for a number of languages. These results indicate that such recognition works very efficiently for candidate generation in spelling correction for many European languages such as English, Dutch, French, German, Italian (and others) with very large word lists of root and inflected forms (some containing well over 200,000 forms), generating all candidate solutions within 10 to 45 milliseconds (with edit distance 1) on a SparcStation 10/41. For spelling correction in Turkish, error-tolerant