An Architecture for Detecting Events in Real-Time using Massive Heterogeneous Data Sources

August 12, 2013
Authors: George Valkanas, Dimitrios Gunopulos, Ioannis Boutsis, Vana Kalogeraki
Conference: BigMine 2013 @SIGKDD 2013

In this work we present a distributed architecture for processing large volumes of
data, that can be used to identify events, in various ways. The first one is to
analyze the Twitter stream as it arrives, whereas the other one is to analyze
communities. Both of them are manipulated in real-time, and we present our
distributed architecture for combining the processing of these two cases. The goal
of this work is to present the architecture of a framework designed to gather,
aggregate and process a wide range of sensory input coming from very different
sources, such as GPS signals, text, acceleration sensors, etc. An additional,
distinctive characteristic of our framework is the active involvement of citizens.