Publication Type
Conference Paper
Version
submittedVersion
Publication Date
11-2018
Abstract
Social networks generate a massive amount of interaction data among users in the form of streams. To facilitate social network users to consume the continuously generated stream and identify preferred viral social contents, we present a real-time monitoring system called River to track a small set of influential social contents from high-speed streams in this demo. River has four novel features which distinguish itself from existing social monitoring systems: (1) River extracts a set of contents which collectively have the most significant influence coverage while reducing the influence overlaps; (2) River is topic-based and monitors the contents which are relevant to users' preferences; (3) River is location-aware, i.e., it enables user influence query on the contents falling into the region of interests; and (4) River employs a novel sparse influential checkpoint (SIC) index to support efficient updates against the streaming rates of real-world social networks in real-time.
Discipline
Databases and Information Systems
Research Areas
Data Science and Engineering
Publication
IEEE International Conference on Data Mining
Publisher
World Scientific Publishing
City or Country
Singapore
Citation
SHA, Mo; LI, Yuchen; WANG, Yanhao; GUO, Wentian; and TAN, Kian-Lee.
River: A real-time influence monitoring system on social media stream. (2018). IEEE International Conference on Data Mining.
Available at: https://ink.library.smu.edu.sg/sis_research/4216
Creative Commons License
This work is licensed under a Creative Commons Attribution-NonCommercial-No Derivative Works 4.0 International License.