Document Type

Conference Proceeding

Publication Date

8-28-2020

Publication Title

2020 IEEE Sixth International Conference on Big Data Computing Service and Applications (BigDataService)

Editor

Lisa O’Conner

First Page

46

Last Page

53

Abstract

Online Social Networks allow users to share experiences with friends and relatives, make announcements, find news and jobs, and more. Several have user bases that number in the hundred of millions and even billions. Very often many users belong to multiple social networks at the same time under possibly different user names. Identifying a user from one social network on another social network gives information about a user's behavior on each platform, which in turn can help companies perform graph mining tasks, such as community detection and link prediction. The process of identifying or aligning users in multiple networks is called network alignment. These similar (or same) users on different networks are called anchor nodes and the edges between them are called anchor links. The network alignment problem aims at finding these anchor links. In this work we propose two supervised algorithms and one unsupervised algorithm using thresholds. All these algorithms use local structural graph features of users and some of them use additional information about the users. We present the performance of our models in various settings using experiments based on Foursquare-Twitter and Facebook-Twitter data (User Identity Linkage Dataset). We show that our approaches perform well even when we use the neighborhood of the users only, and the accuracy improves even more given additional information about a user, such as the username and the profile image. We further show that our best approaches perform better at the HR@1 task than unsupervised and semi-supervised factoid embedding approaches considered earlier for these datasets.

Comments

This is a post-peer-review, pre-copyedit version of a chapter published in the 2020 IEEE Sixth International Conference on Big Data Computing Service and Applications (BigDataService). The final authenticated version is available online at: https://doi.org/10.1109/BigDataService49289.2020.00015.© 2020 IEEE. Personal use of this material is permitted. Permission from IEEE must be obtained for all other uses, in any current or future media, including reprinting/republishing this material for advertising or promotional purposes, creating new collective works, for resale or redistribution to servers or lists, or reuse of any copyrighted component of this work in other works.

SJSU users: Use the following link to login and access the article via SJSU databases.

COinS