A pruning algorithm for Meta-blocking based on cumulative weight

2nd Annual International Conference on Information System and Artificial Intelligence

物理学;计算机科学

Zhang, Fulin^1 ; Gao, Zhipeng^1 ; Niu, Kun^2

State Key Laboratory of Networking and Switching Technology, Beijing University of Posts and Telecommunications, Beijing

100876, China^1

School of Software Engineering, Beijing University of Posts and Telecommunications, Beijing

100876, China^2

关键词: Blocking method;    Blocking process;    Cumulative weight;    Entity resolutions;    Heterogeneous information;    Large datasets;    Pruning algorithms;    Quadratic complexity;   

Others  :  https://iopscience.iop.org/article/10.1088/1742-6596/887/1/012058/pdf
DOI  :  10.1088/1742-6596/887/1/012058

学科分类：计算机科学（综合）

来源: IOP

【 摘 要 】

Entity Resolution is an important process in data cleaning and data integration. It usually employs a blocking method to avoid the quadratic complexity work when scales to large data sets. Meta-blocking can perform better in the context of highly heterogeneous information spaces. Yet, its precision and efficiency still have room to improve. In this paper, we present a new pruning algorithm for Meta-Blocking. It can achieve a higher precision than the existing WEP algorithm at a small cost of recall. In addition, can reduce the runtime of the blocking process. We evaluate our proposed method over five real-world data sets.