Having introduced the theory of Chinese word segmentation algorithm,this paper proposes to improve"the back one word combination"by analyzing the categories of word segmentation ambiguity and its inevitability in order to eliminate it and enhance the precision of segmentation while maintaining the segmentation rate. As a result, it lays solid foundation for search engines to establish indexes.%介绍中文分词算法的理论知识,通过介绍歧义存在的种类,分析分词结果出现歧义的必然性.提出改进"退一字组合法",实现歧义消除.在保持切分速度的前提下,提高切分的精度.为搜索引擎建立索引奠定良好的基础.
展开▼