Abstract:

Sentence component analysis is important and difficult in natural language processing. We first discuss component division problems of single sentences in modern Uyghur, including the relationship between modern Uyghur phrases and sentences and the relationship among the parts of speech.  We then discuss the pretreatment of the modern Uyghur corpus, phrase tag sets, the basic idea of dividing sentence elements, and sentence component analysis algorithm. We further explore a recognition design for identifying modern Uyghur predicate elements, other sentence components, and automatical defining  prediction algorithms to solve the problems of modern Uighur statement analysis and experiment data analysis.

Key words: information processing;Uyghur words;phrase structure;sentence component