Social media has become indispensable for many in modern times and the use of these platforms continues to grow exponentially. Despite the popular use of social networks, most social media platforms do not provide information regarding the propagation of posts in relation to individual users. For example, automated activities on social media create opportunities for manipulation, misinformation, and distrust. Post-specific diffusion network elucidates the who-saw-from-whom paths of a post on social media that are currently not publicly available. A diffusion network for a specific post can reveal trustworthy and/or incentivized connections among users. Unfortunately, such a network is not observable from available social media platform information, and without an analytics tool that tracks the propagation of social media posts, crucial information regarding diffusion paths between users is lost.
Researchers from the University of New Mexico have designed a novel algorithm which identifies the maximum likely diffusion network of a social media post. The algorithm can scale up to thousands of shares from a single post by analyzing inferred diffusion networks and showing discernible differences in information diffusion within various user groups, as well as across local communities. With this in mind, the algorithm can be implemented as a near real-time analytics tool. Discovering differences in inferred networks, showing the disproportionate presence of automated bots, is a potential way to measure the true impact of a post. In addition, the algorithm exploits a myriad set of information that includes temporal, textual, posting history, and social network of the author and the retweeters – a much broader set of data compared to existing methodologies. Effectiveness has been demonstrated over a variety of datasets, both synthetic and real.
- Near real-time tracking of individual posts
- Infers diffusion of a single post
- Explains how a post is being shared
- Monitors social medial platforms
- Exhibits effectiveness toward real and synthetic datasets
- Social Media
Name: Andrew Roerick