摘要
本文采用规则和模板相结合的策略,提出了一种维吾尔人名消歧的方法。该方法首先采用一系列规则,对维吾尔人名进行初步消歧,其次通过消歧模板,对初步消歧的结果进行进一步筛选。在之前做好的维吾尔人名识别系统的基础上做试验,实验结果表明,该方法提高了准确率和召回率。
In this paper,we use a combination of rules and templates to propose a method of Uyghur name disambiguation.The method firstly adopts a series of rules to make preliminary disambiguation of Uighur names,and secondly,through disambiguation templates,the results of preliminary disambiguation are further screened.Based on the well-known Uyghur name recognition system,the experiment results show that the method improves the accuracy and recall rate.
出处
《数字技术与应用》
2017年第12期93-94,共2页
Digital Technology & Application
基金
新疆维吾尔自治区自然科学基金项目(2016D01B010)
新疆高校科研计划青年项目(XJEDU2016S076)
喀什大学校内科研课题资助项目((16)2601)
关键词
维吾尔人名
人名消歧
机器翻译
命名实体识别
Uyghur name
name disambiguation
machine translation
named entity recognition
