A Community Sensing Approach for User Identity Linkage
A Community Sensing Approach for User Identity Linkage
复制标题
DOI：
10.1007/978-3-030-39878-1_18
复制
发表时间：
2019-06
期刊：
影响因子：
--
通讯作者：
Zexuan Wang;Teruaki Hayashi;Y. Ohsawa
中科院分区：
文献类型：
--
作者：
Zexuan Wang;Teruaki Hayashi;Y. Ohsawa
这是从JSAI2019精选的一篇论文的延伸。用户身份链接(UIL，User Identity Linkage)的目标是在不同的在线社交网络中发现相同的个人或实体，这是孤立网络之间的信息传播和不同域之间的信息传递的关键步骤。虽然在这一重要问题上已经提出了许多结对的用户链接方法，但网络中自然存在的社区信息在链接过程中往往被丢弃。在本文中，我们提出了一种新的基于嵌入的方法，通过在单个损失函数中联合优化个体相似度和社区相似度来考虑和利用它们。在从Foursquare和Twitter获得的真实数据集上进行的实验表明，该方法的性能优于UIL中其他常用的基线，后者只考虑用户或实体之间的个体相似性。
This is an extension from a selected paper from JSAI2019. User Identity Linkage (UIL) aims to detect the same individual or entity across different Online Social Networks, which is a crucial step for information diffusion among isolated networks and information transfer between different domains. While many pair-wise user linking methods have been proposed on this important topic, the community information naturally exists in the network is often discarded during process. In this paper, we proposed a novel embedding-based approach that considers and utilizes both individual similarity and community similarity by jointly optimizing them in a single loss function. Experiments conducted on real datasets obtained from Foursquare and Twitter illustrate that the proposed method outperforms other commonly used baselines in UIL, which only consider the individual similarity between users or entities.