All Dates/Times are Australian Eastern Standard Time (AEST)

Technical Program

Paper Detail

Paper IDD4-S5-T3.3
Paper Title Attributed Graph Alignment
Authors Ning Zhang, University of British Columbia, Canada; Weina Wang, Carnegie Mellon University, United States; Lele Wang, University of British Columbia, Canada
Session D4-S5-T3: Network Inference
Chaired Session: Thursday, 15 July, 23:20 - 23:40
Engagement Session: Thursday, 15 July, 23:40 - 00:00
Abstract Motivated by various data science applications including de-anonymizing user identities in social networks, we consider the graph alignment problem, where the goal is to identify the vertex/user correspondence between two correlated graphs. Existing work mostly recovers the correspondence by exploiting the user-user connections. However, in many real-world applications, additional information about the users, such as user profiles, might be publicly available. In this paper, we introduce the attributed graph alignment problem, where additional user information, referred to as attributes, is incorporated to assist graph alignment. We establish sufficient and necessary conditions for recovering vertex correspondence exactly, where the conditions match for a wide range of practical regimes. Our results recover existing tight information-theoretic limits for models where only the user-user connections are available, and further span the full spectrum between these models and models where only attribute information is available.