An integrated information lifecycle management framework for exploiting social network data to identify dynamic large crowd concentration events in smart cities applications

Abstract

With the current availability of an extreme diversity of data sources and services, emerging from the Internet of Things and Cloud domains, the challenge is shifted towards identifying intelligent, abstracted and adaptive ways of correlating and combining the various levels of information. The purpose of this work is to demonstrate such a combination, on one hand at the service level, through integrating smart cities platforms for user level data, and on the other hand at Complex Event Processing, Storage and Analytics capabilities together with Twitter data. The final goal is to identify events of interest to the user such as Large Crowd Concentration (LCC) in a given area, in order to enrich application level information with related event identification that can enable more sophisticated actions on behalf of that user. The identification is based on observation of Twitter activity peaks compared to historical data on a dynamic time and location of interest. The approach is validated through a two-month experiment in the city of Madrid, identifying LCCs in sporting events around two sports venues and analyzing various approaches with relation to the needed thresholds definition.