TY - GEN
T1 - Oases
T2 - 11th IEEE International Conference on Cloud Computing, CLOUD 2018
AU - Xu, Hailu
AU - Hu, Liting
AU - Liu, Pinchao
AU - Xiao, Yao
AU - Wang, Wentao
AU - Dayal, Jai
AU - Wang, Qingyang
AU - Tang, Yuzhe
N1 - Funding Information:
In this paper, we present the online scalable spam detection system (Oases), a distributed and scalable system which detecting the social network spam in an online fashion. By periodically updating the trained classifier through a decentralized DHT-based tree overlay, Oases can effectively harvest and uncover deceptive online spam posts from social communities. Besides, Oases actively filters out new spam and updates the classifiers to all distributed leaf agents in a scalable way. Our large-scale experiments using real-world Twitterdata demonstrate scalability, attractive load-balancing, and graceful efficiency in online spam detection. Future work on Oases will go beyond additional implementation steps, e.g., to implement new specification/configuration APIs for end users, achieve high-availabilityby exploring checkpointing/failover approaches, reduce runtime overhead, all with goals of achieving both good performance and high resource efficiency for large-scale online spam detection. ACKNOWLEDGMENT We gratefully thank the anonymous reviewers for their feedback and thank Florida International University School of Computing & Information Science and GPSC for the support to present this work.
Publisher Copyright:
© 2018 IEEE.
PY - 2018/9/7
Y1 - 2018/9/7
N2 - Web-based social networks enable new community-based opportunities for participants to engage, share their thoughts, and interact with each other. Theses related activities such as searching and advertising are threatened by spammers, content polluters, and malware disseminators. We propose a scalable spam detection system, termed Oases, for uncovering social spam in social networks using an online and scalable approach. The novelty of our design lies in two key components: (1) a decentralized DHT-based tree overlay deployment for harvesting and uncovering deceptive spam from social communities; and (2) a progressive aggregation tree for aggregating the properties of these spam posts for creating new spam classifiers to actively filter out new spam. We design and implement the prototype of Oases and discuss the design considerations of the proposed approach. Our large-scale experiments using real-world Twitter data demonstrate scalability, attractive load-balancing, and graceful efficiency in online spam detection for social networks.
AB - Web-based social networks enable new community-based opportunities for participants to engage, share their thoughts, and interact with each other. Theses related activities such as searching and advertising are threatened by spammers, content polluters, and malware disseminators. We propose a scalable spam detection system, termed Oases, for uncovering social spam in social networks using an online and scalable approach. The novelty of our design lies in two key components: (1) a decentralized DHT-based tree overlay deployment for harvesting and uncovering deceptive spam from social communities; and (2) a progressive aggregation tree for aggregating the properties of these spam posts for creating new spam classifiers to actively filter out new spam. We design and implement the prototype of Oases and discuss the design considerations of the proposed approach. Our large-scale experiments using real-world Twitter data demonstrate scalability, attractive load-balancing, and graceful efficiency in online spam detection for social networks.
KW - DHT based overlay
KW - Online social networks
KW - Spam detection
UR - http://www.scopus.com/inward/record.url?scp=85057446615&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=85057446615&partnerID=8YFLogxK
U2 - 10.1109/CLOUD.2018.00020
DO - 10.1109/CLOUD.2018.00020
M3 - Conference contribution
AN - SCOPUS:85057446615
T3 - IEEE International Conference on Cloud Computing, CLOUD
SP - 98
EP - 105
BT - Proceedings - 2018 IEEE International Conference on Cloud Computing, CLOUD 2018 - Part of the 2018 IEEE World Congress on Services
PB - IEEE Computer Society
Y2 - 2 July 2018 through 7 July 2018
ER -