A Bayesian Hybrid Method for Context-sensitive Spelling Correction

A Bayesian hybrid method for context-sensitive spelling correction Andrew It. Golding Mitsubishi Electric Research Labs 201 Broadway, 8th Floor Cambridge, MA 02139 golding©merl, tom Abstract Two classes of methods have been shown to be useful for resolving lexical ambiguity. The first relies on tile presence of particular words within some distance of tile ambiguous target word; the second uses the pattern of words and part-of-speech tags around the target word. These methods have complementary coverage: the former captures the lexical "atmosphere" (discourse topic, tense, etc.), while tile latter captures local syntax. Yarowsky has exploited this complementarity by combining the two methods using decision lists. The idea is to pool the evidence provided by the component methods, and to then solve a target problem by applying the single strongest piece of evidence, whatever type it happens to be. This paper takes Yarowsky's work as a starting point, applying decision lists to the problem of context-sensitive spelling correction. Decision lists are found, by and large, to outperform either component method. However, it is found that further improvements can be obtained by taking into account not just the single strongest piece of evidence, but all the available evidence. A new hybrid method, based on Bayesian classifiers, is presented for doing this, and its performance improvements are demonstrated. 1 Introduction Two classes of methods have been shown useful for resolving lexical ambiguity. The first tests for the presence of particular context words within a certain distance of the ambiguous target word. The second tests for collocations -- patterns of words and part-of-speech tags around the target word. The context-word and collocation methods have complementary coverage: the former captures the lexical "atmosphere" (discourse topic, tense, etc.), while the latter captures local syntax. Yarowsky [1994] has exploited this complementarity by combining the two methods using decision lists. The idea is to pool the evidence provided by the component methods, and to then solve a target problem by applying the single strongest piece of evidence, whatever type it happens to be. Yarowsky applied his method to the task of restoring missing accents in Spanish and French, and found that it outperformed both the method based on context words, and one based on local syntax. This paper takes Yarowsky's method as a starting point, and hypothesizes that further improvements can be obtained by taking into account not only the single strongest piece of evidence, but all the available evidence. A method is presented for doing this, based on Bayesian classifiers. The work reported here was applied not to accent restoration, but to a related lexical disam- biguation task: context-sensitive spelling correction. The task is to fix spelling errors that happen to result in valid words in the lexicon; for example: I'd like the chocolate cake for ,desert. where dessert was misspelled as desert. This goes beyond the capabilities of conventional spell checkers, which can only detect errors that result in non-words. 39 We start by applying a very simple method to the task, to serve as a baseline …

Content truncated to 500 words