A graph-based approach for population health analysis using Geo-tagged tweets
- PDF / 951,166 Bytes
- 18 Pages / 439.642 x 666.49 pts Page_size
- 79 Downloads / 176 Views
A graph-based approach for population health analysis using Geo-tagged tweets Hung Nguyen1 · Thin Nguyen2 · Duc Thanh Nguyen3 Received: 7 May 2020 / Revised: 13 August 2020 / Accepted: 6 October 2020 / © Springer Science+Business Media, LLC, part of Springer Nature 2020
Abstract We propose in this work a graph-based approach for automatic public health analysis using social media. In our approach, graphs are created to model the interactions between features and between tweets in social media. We investigated different graph properties and methods in constructing graph-based representations for population health analysis. The proposed approach is applied in two case studies: (1) estimating health indices, and (2) classifying health situation of counties in the US. We evaluate our approach on a dataset including more than one billion tweets collected in three years 2014, 2015, and 2016, and the health surveys from the Behavioral Risk Factor Surveillance System. We conducted realistic and largescale experiments on various textual features and graph-based representations. Experimental results verified the robustness of the proposed approach and its superiority over existing ones in both case studies, confirming the potential of graph-based approach for modeling interactions in social networks for population health analysis. Keywords Graphs · Large-scale computing · Health on the web · Population health · Geo-tagged tweets
1 Introduction Population health measurement reflects the dynamic state of physical, mental, and social well-being of a community [18, 43]. Understanding population health is thus essential for
Duc Thanh Nguyen
[email protected] Hung Nguyen [email protected] Thin Nguyen [email protected] 1
Faculty of IT, Nha Trang University, Nha Trang, Vietnam
2
Applied Artificial Intelligence Institute, Deakin University, Geelong, VIC 3220, Australia
3
School of Information Technology, Deakin University, Geelong, VIC 3220, Australia
Multimedia Tools and Applications
governments to identify health-related concerns and develop strategic healthcare programs for communities. Traditionally, population health data is collected via telephone interviews or postal questionnaires. The benefits of this approach include the simplicity of data collection and the reliability of responses. This is mostly because the questionnaires have been designed by professionals, and the population of interest have been actively and intentionally targeted. Despite these advantages, traditional health surveys exhibit two major limitations: expensive cost and time-consuming. For instance, the budget spent for the Behavioral Risk Factor Surveillance System (BRFSS) survey in Florida, US over 5 years 2011 - 2015 was more than 3.5 million USD,1 and the BRFSS reports in 2017 were typically based on the data collected in or before 2015,2 which, in turn, could lead to delayed public health policy decisions. Social behaviors of a population provide cues for the health status of that population. The challenge here is how to
Data Loading...
