藏文字频统计系统中字构件分解算法

发布时间：2018-11-23 21:06

【摘要】：藏文字频统计是藏文信息处理的基础性工作,通过对藏文字的部件、音节、结构和字的频度与通用度等定量统计与定性分析,为藏文信息处理提供基础数据。藏文字是一种由藏文字构件横向和纵向组合而成的拼音文字,在藏文字频统计中不仅要从整字角度统计分析藏文字频度属性,还要统计分析构成其构件的频度及位置属性。因此,在藏文字频统计系统中要分解构成藏文字的各部件。本文通过开发藏文字频统计系统,利用组合构件库结合藏文文法提出了一种藏文字构件分解算法。经测试,该算法不仅简单易行,而且可以有效地确定出各基本构件的位置特征,已应用于项目藏文字频统计系统。
[Abstract]:The frequency statistics of Tibetan characters is the basic work of Tibetan information processing. Through quantitative statistics and qualitative analysis of components, syllables, structures, frequency and generality of Tibetan characters, this paper provides basic data for Tibetan information processing. Tibetan character is a kind of phonetic character composed of horizontal and vertical components of Tibetan characters. In the statistics of Tibetan characters frequency, not only the frequency attributes of Tibetan characters should be statistically analyzed from the perspective of whole characters, but also the frequency and position attributes of the components should be statistically analyzed. Therefore, in the Tibetan character frequency statistical system, the components that constitute Tibetan characters should be decomposed. In this paper, a component decomposition algorithm of Tibetan characters is proposed by developing the Tibetan character frequency statistics system and combining the combination component library with Tibetan grammar. The test results show that the algorithm is not only simple but also can effectively determine the location characteristics of the basic components. It has been applied to the project Tibetan character frequency statistics system.
【作者单位】：青海师范大学藏文信息研究中心;
【基金】：国家社科基金资助项目(09XYY024) 青海师范大学科研基金资助项目
【分类号】：H214;H08

【参考文献】