Data integration for earthquake disaster using real-world data

Acta Geophysica

article

Tian, Chuanzhao¹  Li, Guoqing¹ 

[1] Institute of Remote Sensing and Digital Earth, Chinese Academy of Sciences;University of Chinese Academy of Sciences

关键词: Data integration;    Earthquake disaster;    Numeric data;    Entity resolution;   

DOI  :  10.1007/s11600-019-00381-4

学科分类：地球科学(综合)

来源: Polska Akademia Nauk * Instytut Geofizyki

【 摘 要 】

The purpose of entity resolution (ER) is to identify records that refer to the same real-world entity from different sources. Most traditional ER studies identify records based on string-based data, so the ER problem relies mostly on string comparison techniques. There is little research on numeric-based data. Traditional ER approaches are widely used in many domains, such as papers, gene sequencing and restaurants, but they have not been used in an earthquake disaster. In this paper, earthquake disaster event information that was collected from different websites is denoted with numeric data. To solve the problem of ER in numeric data, we use the following methods to conduct experiments. First, we treat numbers as strings and use string-based approaches. Second, we use the Euclidean distance to measure the difference between two records. Third, we combine the above two strategies and use a comprehensive approach to measure the distance between the two records. We experimentally evaluate our methods on real datasets that represent earthquake disaster event information. The experimental results show that a comprehensive approach can achieve high performance.

【 授权许可】

Unknown