Chinese Synthetic Words Analysis
Lu Jia · Institutional Repositories DataBase (IRDB) · 2008
Recent years, though several practical Chinese morphological analysis systems have been developed in different institutions around the world, there still remain problems in Chinese word segmation.The lack of internal information of Chinese synthetic words (CSW) has become a crucial problem for Chinese morphological analysis systems, which will face various levels of segmentation needs for upper natural language processing applications being developed.In this thesis, we first define the conceptual differences between Chinese single-morpheme words and Chinese synthetic words, and next categorize Chinese synthetic words into several types according to their internal syntactic relation or morphological structure.Then, after annotating a part of Chinese synthetic words, we find a way to automatically classify Chinese synthetic words into compound word and morphologically derived word categories by using machine learning methods.We believe that we can construct a Chinese synthetic word dictionary with these kinds of word internal information, which in the end will help improve morphological analysis and out-of-vocabulary (OOV) word detection of Chinese text.