1.Academy of Cyber,； 2.Key Laboratory of Trustworthy Distributed Computing and Service （BUPT）,， Ministry of Education
Abstract： With the rapid development of social networks, people have various virtual identities in social networks. User identity linkage problem that aims to identify various virtual identities of the same natural person is becoming increasingly important. User identity linkage method can unearth some hidden information and form a complete user profile to promote the development of multiple research fields, such as cross-network recommendation, link prediction, information dissemination, etc. Existing user-profile based model and network-structure based user identity linkage model do not consider the influence difference between different users, and the convergence speed is slow. In order to model the influences between users, multi-head attention mechanism is added to network random-walk based user linkage method in this paper. The experimental results show that it can improve the effectiveness of social network user identity linkage method and improve training efficiency.
Key words : graph embedding；user identity linkage,；multi-head attention mechanism
引言
根據(jù)中國(guó)互聯(lián)網(wǎng)信息中心發(fā)布的第51次《中國(guó)互聯(lián)網(wǎng)絡(luò)發(fā)展?fàn)顩r統(tǒng)計(jì)報(bào)告》,，我國(guó)的在線社交網(wǎng)絡(luò)數(shù)量已經(jīng)增長(zhǎng)到10.67億，互聯(lián)網(wǎng)的普及率也達(dá)到了75.6%[1],。社交網(wǎng)絡(luò)已經(jīng)成為人們?nèi)粘Ｉ钪胁豢苫蛉钡纳缃还ぞ?，抖音、微信,、微博,、X（Twitter）等社交網(wǎng)絡(luò)層出不窮，人們?cè)谏缃黄脚_(tái)上擁有越來(lái)越多的虛擬身份,。根據(jù)全球網(wǎng)絡(luò)指數(shù)（Global Web Index，GWI)發(fā)布的《2019年社交媒體趨勢(shì)報(bào)告》[2],，平均每個(gè)互聯(lián)網(wǎng)用戶擁有的社交網(wǎng)絡(luò)賬號(hào)已經(jīng)從2015年的約6.2個(gè)上升到2019年的近8個(gè),。因此，社交網(wǎng)絡(luò)的用戶身份鏈接（User Identity Linkage）問(wèn)題成為近年來(lái)的研究熱點(diǎn),，為跨平臺(tái)的用戶畫像[3],、虛假身份信息監(jiān)測(cè)[4]、社交網(wǎng)絡(luò)朋友推薦[5],、信息傳播[6],、鏈接預(yù)測(cè)[7]、網(wǎng)絡(luò)動(dòng)力學(xué)分析[8]等很多下游任務(wù)提供了新的研究思路。
