the performance of the existing word segmenters for dealing with this type of ambiguity is still not satisfactory . in the paper, cross ambiguities in chinese running texts are described quantitatively and systematically based on the basis of observations from a chinese corpus with 100m chinese characters 本文根据一个1亿字的大型汉语语料库和一个包含11万词的汉语词表,对交集型歧义切分字段进行了穷举式的调查以及多角度、多层次的统计分类。