These two approaches can be
summarized as follows:
Trigram analysis is a method used to correct spelling by analyzing parts of words
in three-letter sequences. It identifies three possible candidates for representing
meaningful words. N-gram is used to perform trigram analysis.
??? An n-gram is an n-letter sequence, extracted from a word, where n is usually 1,
2, or 3. In general, n-gram analysis techniques check each n-gram in an input
string against a precompiled table of n-gram statistics to determine whether
the n-gram can occur in a word. If it does, its frequency of occurrence in the
words of the language is computed. Strings containing n-grams that do not
occur in words or occur very infrequently are considered to be possible misspellings.
For instance, the word HEPATITIS contains the trigrams: -HE, HEP,
EPA, PAT, ATI, TIT, TIS, IS- where hyphen (-) represents the word boundary
(Yannakoudakis & Fawthrop, 1983).
Ontology-Based Spell ng Correct on for Search ng Med cal Informat on
Copyright ?© 2007, Idea Group Inc. Copying or distributing in print or electronic forms without written permission
of Idea Group Inc.
Pages:
763
764
765
766
767
768
769
770
771
772
773
774
775
776
777
778
779
780
781
782
783
784
785
786
787