机器翻译中模式匹配算法的研究

Research on Pattern Matching Algorithm for Machine Translation

将经验主义方法与传统的基于规则的理性主义体系结合起来，提出了一种面向实例、基于模式的机器翻译方法。在基于模式的翻译中，模式匹配机制对翻译的准确率和效率有着很大影响，论证了在基于模式的机器翻译中采用翻泽模式最大匹配机制的可行性，并给出了一个能够快速收敛的通用最大匹配算法。

An example - oriented and pattern-based approach is presented through combining empiricism with traditional rule-based rationalism. The pattern matching mechanism affects the accuracy and efficiency of the MT system.The feasibility to apply the MM mechanism to translation pattern in pattern- based MT is discussed. And a general MM algorithm that can converge quickly is established.