【作者】 郑逢斌； 姜保庆； 郭拯危； 邱小平； 乔保军；

【Author】 ZHENG Fengbin JIANG Baoqing GUO Zhengwei QIU Xiaoping QIAO Baojun(Intelligence Control Development Center, Southwest Jiaotong University, Chengdu 610031, China); (Cottege of Computer and Information Engineering, Henan University, Kaifeng 475001, China)

【机构】 西南交通大学智能控制开发中心； 河南大学计算机与信息工程学院；

【摘要】 本文介绍了应用于报刊书籍政治性错误校对的一个语义校对软件系统的设计原理和实现过程。本系统先把文本切分成词语,找出敏感词,合并成术语,部分术语通过短语本身或左右片段即可做出正误判断,另一部分术语要用整个句子语义与知识库进行模糊匹配才能得到它的错误隶属度.更多还原

【Abstract】 In this paper, the design principle and execution process of the new semantic collation system about political error in newspapers and books are introduced Following contents are included in the system: first, sentences are cut into words; second, sensitive terms are picked up. For these sensitive terms, the correctness of some can be judged by themselves or near segments, whereas for others, their error membership can only be obtained by fuzzy matching between the sentences they belong to and that of knowledge database.更多还原