Users Connection across Social Media Sites Based On Users'Relationship Vector

DOI

10.2991/cnct-16.2017.19How to use a DOI?

Keywords

Online social networks, Multiple identities, Machine learning, Feature vector

Abstract

The identification and association of multiple identities in different online social networks (osns) is an important problem, and also is the basis for many applications. At present, most of technologies try to solve this problem by matching the username of social networks or calculating the similarity of a pair of users' personal information from different platforms. However, due to the anonymity of social networks, these methods often fail to identify and associate multiple virtual identities. In this paper, we propose a classification method based on machine learning. Our method jointly consider the time, the text and the topic of the similarity to construct the feature vector to characterize the user's relationship. And we use the feature vectors to train the classifier. The model is evaluated on real world dataset, the twitter and sina weibo. The experimental results show that our method is effective.

Copyright

© 2017, the Authors. Published by Atlantis Press.

Open Access

This is an open access article distributed under the CC BY-NC license (http://creativecommons.org/licenses/by-nc/4.0/).

Download article (PDF)

TY  - CONF
AU  - Zhou YAN
AU  - Shu-dong LI
AU  - Wei-hong HAN
AU  - Bin ZHOU
AU  - Wen-xiang HAN
PY  - 2016/12
DA  - 2016/12
TI  - Users Connection across Social Media Sites Based On Users'Relationship Vector
BT  - Proceedings of the International Conference on Computer Networks and Communication Technology (CNCT 2016)
PB  - Atlantis Press
SP  - 137
EP  - 147
SN  - 2352-538X
UR  - https://doi.org/10.2991/cnct-16.2017.19
DO  - 10.2991/cnct-16.2017.19
ID  - YAN2016/12
ER  -