Biology Reference
In-Depth Information
Breaking News
Warning
Old News
Context
Not Disease Related
Acquiring
Categorizing
Clustering
Filtering
Figure 6.2
Stages of HealthMap surveillance. (1) Web-based data are acquired from a variety of Web sites
every hour, 7 days a week (ranging from rumors on discussion sites to news media to validated
official reports). (2) The extracted articles are then categorized by pathogen and location of
the outbreak in question. (3) Articles are then analyzed for duplication and content. Duplicate
articles are removed; while those that discuss new information about an ongoing situation are
integrated with other related articles and added to the interactive map. (4) Once classified, arti-
cles are filtered by their relevance into five categories. Only articles tagged as “breaking news“
or “warning“ are added as markers to the map. (From Brownstein, J. et al. 2008. Surveillance
Sans Frontieres: Internet-Based Emerging Infectious Disease Intelligence and the HealthMap
Project.
PLoS Med
5 (7):6. With permission.)
past outbreak), (4) context (information on disease context, e.g., preparedness
planning), and (5) not disease-related (information not relating to any disease
[3-5 are hidden from the main HealthMap user interface]) (Brownstein et al.
2008; Keller et al. 2009). Another category recently added is applied to alerts that
may constitute a public health event of international significance per the WHO
Interim Guidance for the use of Annex 2 of the International Health Regulations
(2005) (WHO 2008). Alerts tagged as internationally significant are automati-
cally e-mailed to HealthMap users who have opted to receive these notifications.
Finally, once the automated process is complete, an analyst corrects misclassifi-
cations where necessary (Brownstein et al. 2008).
6.3 Human Analysis
Currently, only one analyst works to review and correct the posts; however,
additional resources would allow for more detailed multilingual curation of
collected reports (Brownstein et al. 2008). Misclassifications such as incorrect
tagging (e.g., “breaking” versus “context”), incorrect geographic location,
and/or incorrect disease classifications occur occasionally, and are manually
