The Grammatical Function Of The Unknown Word Guessing

In the actual text, the occurrence of the unknown words is frequent and unpredictable, which affects the correctness of the automatic syntax analyses. The uncertainty of the unknown words brings the uncertainty of the grammar function. The method of allocating the probability value to each unknown word in average always generates multiple syntax trees during the automatic syntax analyses and overloads the automatic syntax analyses.This paper is to guess the syntax functions of the unknown words to reduce the number of the generated syntax analyses. The word-building mode is used for this. In this paper, I limit the productive word-building modes according to the Modern Chinese Dictionary (2005) and the large-scale corpus of 20,000,000 words, and give the quantitative description on the syntax functions for each word-building mode with this large-scale corpus. A table of the syntax functions of the word-building modes is prepared. This table gives the syntax function collection of each word-building mode and the probability of each syntax function. If an unknown word matches a word-building mode successfully, the syntax function collection of each word-building mode and the probability of each syntax function for this unknown word are given. The object to be studied in this paper is the unknown word of the morphology word. Additionally, the lexicon word and the morphology word are differentiated. A table of the morphology words is prepared according to the Modern Chinese Dictionary.