基于Hash方法的机器翻译词典的组织与构造

给出了一种机器翻译词典的二级Ｈａｓｈ方法，并用此方法组织与构造了中日机器翻译标准词典，该方法可通过统计以各种字集开头的词的数目来规划数据块的分配，同时又将统计结果作为Ｈａｓｈ方法的参数来调节数据块的分配，这种调节作用非常有利于控制分布的均匀性，减少冲突，提高空间利用率，该方法不仅免去了常规的以索引方式组织词典模式中对各级索引的繁琐的维护工作，而且也提高了分词效率。

Machine translation dictionary based on Hash method

This paper presents a two level hash method to organize and build the machine translation dictionary. This method can dynamically organize the data blocks' distribution through the statistics, and use the result of statistics to distribute the data blocks in the dictionary file. This dynamic adjustment gives the dictionary a lot of advantage. It makes the data blocks' distribution even decrease the overflow and increase the space usage rate. And also it need not maintain the indexes of ordinary dictionary model with indexes. This method increases the efficiency in word cutting.