The structure was classified in detail using the set theory, and we acquired a series of reasons that producing Combinatorial ambiguity and segment variation.

An inconsistent result about the combinatorial ambiguity causes a discussion between the definitions of combinatorial ambiguity and segmentation variation.

Resolving combinational ambiguity in Chinese word segmentation based on rule mining and Nave Bayes method;

Aiming at cross ambiguity and combinational ambiguity,using the HowNet as the main semantic resources,using knowledge map s knowledge as representation,using the algorithms mentioned in this paper based on semantic calculations,and calculating the discrepancy field of semantic context,to select the correct sentence cut programs,so as to achieve the objective to remove ambiguities.

On the foundations of modern Chinese corpus processing principles and certain type time word segmentation ambiguity,this paper proposes a statistical language model and corresponding approach based on maximum likelihood to solve the ambiguous problem,and it reaches a 90% accuracy which shows the eff.

A Chinese words segmentation compare software is introduced which includes the architecture s design to hold all kinds of algorithm, manual segmenting ambiguity partition, display statistical result.