Using Twitter for Public Health Surveillance from Monitoring and Prediction to Public Response

Abstract

Twitter is a social media platform where over 500 million people worldwide publish their ideas and discuss diverse topics, including their health conditions and public health events. Twitter has proved to be an important source of health-related information on the Internet, given the amount of information that is shared by both citizens and official sources. Twitter provides researchers with a real-time source of public health information on a global scale, and can be very important in public health research. Classifying Twitter data into topics or categories is helpful to better understand how users react and communicate. A literature review is presented on the use of mining Twitter data or similar short-text datasets for public health applications. Each method is analyzed for ways to use Twitter data in public health surveillance. Papers in which Twitter content was classified according to users or tweets for better surveillance of public health were selected for review. Only papers published between 2010–2017 were considered. The reviewed publications are distinguished by the methods that were used to categorize the Twitter content in different ways. While comparing studies is difficult due to the number of different methods that have been used for applying Twitter and interpreting data, this state-of-the-art review demonstrates the vast potential of utilizing Twitter for public health surveillance purposes.

Google Health Trends performance reflecting dengue incidence for the Brazilian states

“…As a consequence, the exploitation of Internet data as a source to characterize epidemiological patterns for communicable and non-communicable diseases has been promoted since the mid-90's under the concept of digital epidemiology [28][29][30][31]. These efforts have focused on leveraging freely available information from Twitter, Google, Wikipedia, among others, to follow traces of disease patterns in the population [32][33][34]. Following the pioneering work of Eysenbach G. on using web-based search queries to track influenza [35] and other efforts that used Google-derived data for influenza in the United States [36,37] and dengue in different countries [38,39], Google developed Google Flu Trends (GFT) in 2009 and Google Dengue Trends (GDT) in 2011, as specific disease surveillance tools for digital epidemiology.…”

Text Mining in Big Data Analytics

“…Moreover, text mining professionals are increasingly becoming high in demand. Furthermore, text mining may have the power to deliver significant insights to society and individuals, especially with respect to public health [258,259], healthcare [260,261], and education [262][263][264][265], and help evaluate social issues, such as crime (including cybercrime) [245,266,267], child abuse [268], and poverty [269]. Nevertheless, actions must be taken in time to efficiently solve the legal, ethical, and privacy concerns contained in the use of personal data.…”

Risk assessment strategies for early detection and prediction of infectious disease outbreaks associated with climate change

“…The text, numbers and dates are not organized in a data model, such as a database, that can be used for automated event detection and risk modelling; therefore, open-source data must be processed to extract and structure information about what happened, where it happened, when it happened and to whom it happened. The EBS systems use natural language processing (NLP) methods to process and understand event narratives (46)(47)(48). Natural language processing is a field of research dedicated to understanding human discourse (49).…”