论文部分内容阅读
在对目前各种作者重名消解方法进行总结的基础上,针对中文文献题录数据特征,将重名消解问题转换为同名作者文献的分类问题,提出一种基于规则和相似度的重名消解框架模型,并对其中的分解规则和合并规则进行详细的算法描述,最后选取3个学科的重名作者数据集进行实验,实验结果表明该模型能有效提高作者重名消解的准确率。
On the basis of summarizing the current author’s method of eliminating the duplicate names, aiming at the characteristics of the data of the Chinese bibliographic data, the problem of resolving the duplicate names into the classification of the author’s authors of the same name is proposed, and a duplicate name resolution based on the rules and similarities is proposed Frame model. The algorithm is described in detail in the disassembly rules and the merge rules. Finally, the data set of the author’s author in three disciplines is selected for experiment. The experimental results show that the model can effectively improve the accuracy of the author’s name resolution.