一个并行分词体系结构模型

A System Architecture Model of Parallel Word Segmentation

文章提出并讨论了一种并行全切分分词方法,给出了该方法在MIMD_SM上的并行分词体系结构模型。经过单CPU机上的模拟实验,证明该模型是可行而且有效的。该模型为克服串行全切分存在的复杂性及效率问题提供了新的思路,对于深入研究书面汉语的并行分词问题及实现具有实际意义。

A method of parallel omni-word segmentation for written Chinese is proposed and discussed in this paper,and its system architecture model on MIMD_SM is given.Its simulation on a single CPU hardware has proved that it is feasible and effective.This model gives a new way to overcome the serial omni-word segmentation's disadvantage of high complexity and low efficiency,and it is conducive practically to the deep researches and implementation of parallel word omni-segmentation of written Chinese.