Orthographic Disambiguation Incorporating Transliterated Probability

Orthographic variance is a fundamental problem for many natural language processing applications. The Japanese language, in particular, contains many orthographic variants for two main reasons: (1) transliterated words allow many possible spelling variations, and (2) many characters in Japanese nouns can be omitted or substituted. Previous studies have mainly focused on the former problem; in contrast, this study has addressed both problems using the same framework. First, we automatically collected both positive examples (sets of equivalent term pairs) and negative examples (sets of inequivalent term pairs). Then, by using both sets of examples, a support vector machine based classifier determined whether two terms (t1 and t2) were equivalent. To boost accuracy, we added a transliterated probability P (t1|s)P (t2|s), which is the probability that both terms (t1 and t2) were transliterated from the same source term (s), to the machine learning features. Experimental results yielded high levels of accuracy, demonstrating the feasibility of the proposed approach.