基于关联规则提取的外语翻译信息密度检测  

Foreign language translation information density detection based on association rule extraction

作　　者：孙瑞[1] 吕灏楠 SUN Rui;LV Hao-nan(Xi’an Innovation College of Yan’an University,Xi’an 710100,China;University of Edinburgh,Edinburgh EH89JU,U.K)

机构地区：[1]延安大学西安创新学院,西安710100 [2]爱丁堡大学,苏格兰爱丁堡EH89JU

出　　处：《信息技术》2024年第4期83-86,92,共5页Information Technology

基　　金：2021年陕西省哲学社会科学重大理论与现实问题研究年度一般项目(2021ND0430);陕西省自然科学基金(2020JM5246)。

摘　　要：为了提高外语翻译信息密度检测的有效性,提出基于关联规则提取的外语翻译信息密度检测方法。构建语料特征信息挖掘的二元模型,并采用关联规则提取方法计算关联规则分布特征参数,完成语料特征的提取。根据语料特征提取结果,计算信息密度检测中心点与信息增量特征的多维尺度信息。计算信息的平均可达密度,将其作为密度检测结果进行输出。实验结果表明,相较于传统对比方法,所提方法可以提高外语翻译信息密度检测精度,检测精度在97%左右。In order to improve the effectiveness of foreign language translation information density detection,a foreign language translation information density detection method based on association rule extraction is proposed.The corpus feature is extracted by constructing a binary model of corpus feature information mining,and using the association rule extraction method to calculate the distribution feature parameters of association rules.According to the results of corpus feature extraction,the multi-dimensional scale information of information density detection center point and information increment feature is calculated.The average achievable density of the information is calculated and output as the density detection result.The experiment results show that,compared with the traditional comparison method,the proposed method can improve the detection accuracy of foreign language translation information density,and the detection accuracy is more than 97%.

关 键 词：关联规则提取 外语翻译 信息密度检测 信息增量特征 平均可达密度 

分 类 号：TP391[自动化与计算机技术—计算机应用技术]