Mining Atomic Chinese Abbreviation Pairs: A Probabilistic Model for Single Character Word Recovery

An HMM-based Single Character Recovery (SCR) Model is proposed in this paper to extract a large set of “atomic abbreviation pairs”from a large text corpus. By an “atomic abbreviation pair,”it refers to an abbreviated word and its root word (i.e., unabbreviated form) in which the abbreviation is a single Chinese character. This task is interesting since the abbreviation process for Chinese compound words seems to be “compositional”; in other words, one can often decode an abbreviated word, such as “台大”(Taiwan University), character-by-character back to its root form. With a large atomic abbreviation dictionary, one may be able to recover multiple-character abbreviations more easily. With only a few training iterations, the acquisition accuracy of the proposed SCR model achieves 62% and 50 % precision for training set and test set, respectively, from the ASWSC-2001 corpus.