基于标注词典和规则的维吾尔文动词词干提取方法

摘要

In this paper, carried research on the Uyghur verb, on this basis of artificially tagged verb stemming corpus, we studied the verb inflection formalization, and proposed the Uyghur verb categories inflectional suffix attaching frame. On this basis of artificially tagged verb stemming corpus, we collected suffix attaching rules collection, and build artificially corrected suffix attaching rules collection. Finally, we proposed Uyghur Verb stemming method based on artificially tagged verb stemming corpus, suffix attaching rules collection and verb categories inflectional suffix attaching frame, the accuracy of this method reached 84.15%.%利用语言学专家人工标注的语料库对维吾尔文动词进行研究，并总结出了维吾尔文动词范畴的连接规则框架．在人工标注的实例库的基础上，收集了词缀连接规则集合，经人工纠正构建了词缀连接规则库集合．最终，结合人工标注实例库、词缀连接规则库及维吾尔文动词范畴的连接规则框架提出了维吾尔文动词词干提取方法，该方法的独立实验准确率达到了84.15%．