… Abstract—The typical aim of UserIdentityLinkage (UIL) is to detect when users from across … span from user-profile-based, through user-generated-content-based, user-behavior-based …
Z Zhong, Y Cao, M Guo, Z Nie - … of the AAAI Conference on Artificial …, 2018 - ojs.aaai.org
… We propose CoLink, a general unsupervised framework for the UserIdentityLinkage problem. CoLink separates the user profile into two independent views: user attributes and …
… the multi-platform useridentitylinkage problem based on a new concept of Latent User Space, which more naturally models the relationship among the underlying real user and the var…
… Once our model achieves good performance on this simple task, we start to train it with the final useridentitylinkage task on different mobility data source. This warm-up task is to …
X Li, Y Cao, Q Li, Y Shang, Y Li, Y Liu, G Xu - World Wide Web, 2021 - Springer
… identities, which could contribute to the subsequent linkages in following matching steps. To address this problem, we transform useridentitylinkage … optimize the linkage strategy from …
W Xie, X Mu, RKW Lee, F Zhu… - 2018 IEEE International …, 2018 - ieeexplore.ieee.org
… novel framework that linksuseridentities across multiple OSNs through the use of a network … For a fair comparison, we introduce more anchor links by linkinguseridentities which share …
J Zhou, J Fan - IEEE INFOCOM 2019-IEEE conference on …, 2019 - ieeexplore.ieee.org
… • We propose a reliable iterative identitylinkage mechanism in TransLink. It can maximally improve the precision and recall rate by iteratively appending the newly connected anchor …
… nodes, we intend to find those nodes which are the same useridentity across networks, namely useridentitylinkage predictions. Here, we only rely on the network topology to learn …
… In this section, we define preliminary concepts used in this paper and the useridentitylinkage problem. Without loss of generality, we focus on useridentitylinkage in two social networks…