Similarity and Edit Distance Algorithms for the Korean Alphabet using One-Dimensional Array of Phonemes
Kangho Roh, Kun-Soo Park, Hwan-Gue Cho, So-Won Chang · 2011
The edit distance problem is finding the minimum number of edit operations to transform a string into another one. There are some algorithms that compute an optimal edit distance for one-dimensional languages such as the English language. However, there was little research to find the edit distance for more complicated languages such as the Korean. In this paper, we define two measures of similarity using one dimensional array of phonemes by extending previous edit-distance algorithms for the Korean alphabet based on the properties of syllables and phonemes, and the phoneme classification system. The algorithms presented in this paper, show better experimental results than the previous algorithms for the data of similar words and Korean slang.