关键词: 维吾尔文本, 分词, 组词, 语义词, 频繁模式

Abstract: It is very easy to get the words in Uighur text lines by the natural delimiters such as spaces, but it is difficult to obtain the completely structured semantic words. Therefore,many kinds of text processing methods always seem not to be very effective. This paper put forward a new concept of Uyghur word grouping and introduced the frequent pattern mining method in data mining scheme, and combined the Uyghur language features, turned the pattern mining problem without prior knowledge into a pattern matching with special pattern, and proposed a fast and efficient frequent pattern mining algorithm to obtain the Uyghur words with complete semantics. The experimental results show that, words obtained by this algorithm are stable in structure, and semantically complete and independent.

Key words: Uyghur text, word segmentation, word grouping, semantic word, frequent pattern