汉字字频统计方法的改进

发布时间：2018-08-03 17:43

【摘要】：字频统计是研究实际使用中汉字价值的一种重要方法。以往的字频统计取得了重要的成果,对汉字规范、汉字教学、汉字字典编纂、汉字信息处理以及汉字理论研究都起了重要作用。但也存在基于语料库统计字频造成字频统计结果失真;统计缺乏统一而明确的统计单位,统计结果不科学等问题。解决问题的关键在于建立满足文字统计要求的字料库,必须明确界定统计单位,划分不同层级的汉字统计单位,根据不同需要对其进行分类分层的字频统计。以印刷楷书为例,可以从描写、规定两个方面区分汉字单位。字的单位确定了,字频统计则可以根据不同的统计目的规定统计单位,便于统计数据的使用以及不同统计数据的比较,使得字频统计的科学性得到保障。
[Abstract]:Word frequency statistics is an important method to study the value of Chinese characters in practical use. In the past, important achievements have been made in the statistics of character frequency, which plays an important role in the standardization of Chinese characters, the teaching of Chinese characters, the compilation of Chinese characters dictionaries, the processing of Chinese characters' information and the study of Chinese characters' theory. However, there are also some problems such as the distortion of the word frequency statistics based on the corpus, the lack of unified and definite statistical units and the unscientific statistical results. The key to solve the problem lies in the establishment of a character database that meets the requirements of character statistics. The statistical units must be clearly defined, the Chinese character statistics units of different levels should be divided, and the word frequency statistics should be classified and stratified according to different needs. Take printing regular script as an example, we can distinguish the Chinese character unit from the description, the stipulation two aspects. The word frequency statistics can be defined according to different statistical purposes, which is convenient for the use of statistical data and the comparison of different statistical data, so that the scientific nature of word frequency statistics can be guaranteed.
【作者单位】：北京师范大学民俗典籍文字研究中心;北京师范大学中国文字整理与规范研究中心;
【分类号】：H12

【共引文献】