A computational model of Korean morphological analysis: A prediction-based approach

In this paper, we present an efficient morphological analysis model for Korean which produces all the feasible sequences of morphemes of a given input word. This model is deterministic in applying spelling rules and works with little computational redundancy in processing complex and ambiguous words. The computational efficiency is made possible by three new techniques: first, a method for interpreting and compiling spelling rules; second, predictive rule application which restricts the spelling rules suitable for the input word; third, morphological chart parsing which enables the analyzer to avoid recomputing analyzed substrings in case the input word is morphologically ambiguous. Our model has been tested with words selected from a corpus of Korean elementary textbooks. Experimental results show that our model guarantees fast and reliable processing.