使用机械分词与特征扫描实现对甲骨卜辞的切分  

Segmentation and Tagging of the Shang Oracle-bone Inscriptions Using Mechanical Word Segmentation and Characteristics Scanning

作　　者：陆公正[1] 开金宇[2] 

机构地区：[1]苏州市职业大学计算机工程学院,江苏苏州215104 [2]安阳师范学院河南省甲骨文信息处理重点实验室培育基地,河南安阳455000

出　　处：《苏州市职业大学学报》2013年第4期2-6,共5页Journal of Suzhou Vocational University

基　　金：国家自然科学基金资助项目(61272259);国家自然科学基金资助项目(6087508);苏州市云计算智能信息处理高技术研究重点实验室开放基金资助项目(SXZ201303)

摘　　要：依据甲骨文语法及甲骨文词典对甲骨卜辞进行切分,是建立甲骨文语料库实现计算机辅助甲骨文考释的前提和基础.针对甲骨卜辞的切分,采用机械分词与特征扫描相结合的中文分词技术对200条比较完整的卜辞进行切分实验,将只使用机械分词和结合了特征扫描的两种方法进行比较.结果表明,采用机械分词与特征扫描相结合,实现甲骨卜辞的切分的正确率和效率都较高.To segment the Shang Oracle-bone Inscriptions correctly based on the grammar and the dictionary of the Shang Oracle-bone Inscriptions is the prerequistite for and basis of the establishment of the corpus of the Shang Oracle-bone Inscriptions to realize computer-aided textual explanation.This paper adopts the technology of modern Chinese word segmentation,a combination of mechanical word segmation and Characteristics scanning. In the experiment,200 pieces of Shang Oracle-bone Inscriptions are the experimental samples.We compare the results of the method combined with and without characteristics scanning.The experimental results show that the accuracy and ef？ciency of segmentation proposed in this paper are both high.

关 键 词：甲骨卜辞 分词技术 机械分词 特征扫描 LUCENE 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]