K-partite graph visualization
This is a video demonstration of the K-partite graph visualization of the VAST-2008 CDR(Call Detail Records) dataset. This dataset provided information regarding the calling activities of 400 unique cell-phones over a period of 10 days. The goal of the task was to discover a set of users that changed phone devices in the last 3 days of the corresponding period.
The k-partite graphs visualization depicts the graph representation of three data dimensions, namely: day of the call, call destination (To), call origin (From). Individual bi-partite graphs depict the graph representation of each data dimension. The analyst uses abstraction to visualize the largest call activity for the 10-day period, in order to find two different sets of users that are active at different time periods. By further focusing only on these two sets of users and depicting their social interactions, it is revealed that multiple groups comprised of 2 users have very similar social behavior.
Application of the NEMESYS Framework for detecting anomalies in mobile cellular networks
Example of hypothesis testing and validation using the NEMESYS Framework
Example of the integrated NEMESYS Framework and the intercommunications between the different modules
Application of the analytics tool developed within NEMESYS for the analysis of data in mobile cellular networks
Application of the analytics tool developed within NEMESYS for the analysis of twitter data from the period of the terrorist attack on Paris (November / 2015)