Automated Social Network Epidemic Data Collector

This paper presents the architecture of a system capable of collecting epidemiological data from social network services, and a preliminary prototype collects data from Twitter into a local database, where it can be analyzed and made available to other applications.

Abstract

Recent epidemiological surveillance projects began collecting data from the Internet to identify inflectional diseases propagation. These systems collect data from pre-selected data sources somehow related to the subject. However, other sources, like web social networks, may present early evidences of an infection event. With the increasing popularity of social networks, where people post personal data interactively, we may hint the outbreak of an epidemic event from these data. This paper presents the architecture of a system capable of collecting epidemiological data from social network services. A preliminary prototype collects data from Twitter into a local database, where it can be analyzed and made available to other applications. Initial evaluation results show that relevant epidemiological data can be found in this source.