Cross-network user alignment method based on deep learning

• Main Authors: TAN JIANLONG, GUO LI, ZHOU CHUAN, CHENG ANFENG, ZHU YUJIA
• Format: Patent
• Language: Chinese; English
• Published: 18.10.2019
• Subjects: CALCULATING; COMPUTER SYSTEMS BASED ON SPECIFIC COMPUTATIONAL MODELS; COMPUTING; COUNTING; ELECTRIC DIGITAL DATA PROCESSING; PHYSICS

The invention discloses a cross-network user alignment method based on deep learning. The method comprises the following steps: 1) training a deep learning framework by utilizing a user pair set witha known label and structure information of user pairs; wherein the user pairs are aligned user pairs, that is, the ith user in the social network GA and the mth user in the social network GB are the same user; and 2) according to the structure information of the social network, extracting the structure information of the to-be-aligned users from the social network GA and the social network GB respectively, inputting the structure information into the trained deep learning framework by taking the user pair as a unit, and determining whether the user pair belongs to the same real person for classification prediction. According to the invention, the accuracy of cross-network user alignment is greatly improved. 本发明公开了一种基于深度学习的跨网络用户对齐方法。本方法为：1)利用已知标签的用户对集合及用户对的结构信息训练深度学习框架；其中为已对齐的用户对，即社交网络G中第i个用户与社交网络G中第m个用户是同一用户；2)根据社