Connecting Chinese Users Across Social Media Sites

DOI

10.2991/ic3me-15.2015.245How to use a DOI?

Keywords

virtual identity; Chinese; username; bigdata; pinyin.

Abstract

The usage of social network usernames in the research of social identity linkage has been proved, especially for the English usernames. However, how to properly connect Chinese user identities by matching the usernames remains to be explored. Since a Chinese user may name or rename his/her usernames in different ways (e.g. using a Chinese username or translating it into English, using simplified Chinese characters or traditional ones, converting some words in a given username to their homophonic words, etc.), it is more difficult to connect Chinese users than connect English users by matching usernames. This paper proposes a kind of language mapping method which can translate different type of Chinese words of a given username into their corresponding Pinyin words. However, the number of user identities in a given social network can be very large, thus the username matching process between two social networks is very costly. Basically, we use the Hadoop and Spark frameworks to conquer the efficiency problems. We also have a study on various username matching algorithms, and figure out the features that are useful in Chinese username matching.

Copyright

© 2015, 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  - Yanan Li
AU  - Junxing Zhu
AU  - Zhongcheng Zhou
AU  - Bin Zhou
AU  - Xiaobo Wu
PY  - 2015/08
DA  - 2015/08
TI  - Connecting Chinese Users Across Social Media Sites
BT  - Proceedings of the 3rd International Conference on Material, Mechanical and Manufacturing Engineering
PB  - Atlantis Press
SP  - 1273
EP  - 1279
SN  - 2352-5401
UR  - https://doi.org/10.2991/ic3me-15.2015.245
DO  - 10.2991/ic3me-15.2015.245
ID  - Li2015/08
ER  -