基于不同分词框架的学术论文翻译系统  

Academic Paper Translation System Based on Different Word Segmentation Frame

作　　者：官新均 陈珂[2] 苏旋 邹权[1,4] 

机构地区：[1]厦门大学信息科学与技术学院,福建厦门361005 [2]广东石油化工学院计算机科学与技术系,广东茂名525000 [3]仟首网络科技有限公司,上海200060 [4]厦门大学深圳研究院,广东深圳518000

出　　处：《厦门大学学报（自然科学版）》2013年第6期781-786,共6页Journal of Xiamen University：Natural Science

基　　金：国家自然科学基金项目(61001013;61102136);福建省自然科学基金项目(2011J05158);深圳市科技创新基础研究项目(JCYJ20120618155655087)

摘　　要：针对国内学者和研究人员在撰写英语论文时出现一些术语对应翻译不正确的情况,设计实现了基于术语词库框架的学术论文例句检索系统.由于该检索系统的查询关键字多为专业术语,如新词或未登录词,很难被普通分词方法有效分词,对此提出了一种自建专业术语词典树的方法,有效地提高了分词的准确率,返回更为精准的检索结果.Since the domestic scholars and researchers were always in trouble with wrong corresponding translation of terminologies when they wrote academic papers,a retrieval system of exemplified sentences from research papers based on terminology dictionary has been designed and realized.However,the key words of this retrieval system are mostly new words or out-of-vocabulary words,which is inefficient by the common word segmentation.A new method constructing terminologies tree is proposed,which increases the accuracy rate of word segmentation and the results are more effecive.

关 键 词：信息检索 机器翻译 分词框架 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]