GMA: An adult account identification algorithm on Sina Weibo using behavioral footprints
GMA: An adult account identification algorithm on Sina Weibo using behavioral footprints
复制标题
GMA：使用行为足迹的新浪微博成人帐户识别算法
DOI：
10.1016/j.future.2017.08.032
复制
发表时间：
2019
期刊：
影响因子：
--
通讯作者：
Rodrigues Joel J P C
中科院分区：
文献类型：
--
作者：
Wang Lei;Niu Jianwei;Rodrigues Joel J P C
The uncontrolled spread of sexually explicit content (i.e., adult content) on online social networks (e.g., Sina Weibo) has become an emerging yet acute challenge. An effective method to identify accounts spreading adult content automatically is of significant values in user experience improvement and children protection. Traditional adult content detection and behavioral characteristics of individual account analysis techniques are ill-suited for identifying adult accounts on Sina Weibo due to the collaborative working of adult groups. In this paper, we propose an adult account identification algorithm, which is termedGMA, by analyzing behavioral features (i.e., behavioral footprints) of both groups and individual accounts. InGMA, a novel relation-based model, which considers the inter-relationships among groups, individual accounts and message sources, is applied to identify adult accounts. The experimental results show that compared with state-of-the-art methods, our proposedGMAalgorithm can improve the performance of adult account identification on Sina Weibo.
