摘要
中文用户生成内容大量产生,其随意性、不规范及含义模糊等特征带来信息组织与利用的难题。文章构建面向中文用户生成内容的关联数据混搭系统模型,利用数据层、查询层、整合层和应用层功能,将豆瓣网电影评论信息与DBPedia数据集、LinkedMDB数据集进行有效关联。实验表明该系统能够利用关联开放数据减少用户生成内容的含混和不确定,帮助用户获取丰富的外链数据。
It is a hard task to organize and use Chinese user generated content because of the features of arbitrariness, lack of standardization and ambiguity. This paper constructed a linked data mashup system for Chinese user generated content. The system included four layers: data layer, query layer, integrated layer and application layer. The authors processed the data about movies from DBPedia, LinkedMDB and Douban as a case study. The results showed it enabled users to reduce uncertainty and ambiguity of user generated content by providing more related results from different linked datasets and helped users to get more rich data from external links.
出处
《图书馆学研究》
CSSCI
2017年第8期51-58,共8页
Research on Library Science
基金
教育部人文社会科学规划基金项目"面向用户生成内容组织的关联数据技术应用研究"(项目编号12YJA870029)
中央高校基本科研业务费专项资金项目"基于关联数据的命名实体消歧研究"(项目编号CCNU14A02014)的研究成果之一