Statistical Models for Deep-structure Disambiguation

In this paper, an integrated score function is proposed to resolve the ambiguity of deepstructure, which includes the cases of constituents and the senses of words. With the integrated score function, different knowledge sources, including part-of-speech, syntax and semantics, are integrated in a uniform formulation. Based on this formulation, different models for case identification and word-sense disambiguation are derived. In the baseline system, the values of parameters are estimated by using the maximum likelihood estimation method. The accuracy rates of 56.3% for parse tree, 77.5% for case and 86.2% for word sense are obtained when the baseline system is tested on a corpus of 800 sentences. Afterwards, to reduce the estimation error caused by the maximum likelihood estimation, the Good-Turing's smoothing method is applied. In addition, a robust discriminative learning algorithm is also derived to minimize the testing set error rate. By applying these algorithms, the accuracy rates of 77% for parse tree, 88,9% for case, and 88.6% for sense are obtained. Compared with the baseline system; 17.4% error reduction rate for sense discrimination, 50.7% for case identification, and 47.4% for parsing accuracy are obtained. These results clearly demonstrate the superiority of the proposed models for deep-structure disambiguation.