Monitoring influenza outbreaks while mining the Social Media
Twitter has been used as real time source for various public health applications. This amount of unsorted data, lead us to the idea of creating a real time application that tracks the flu. Flutrack platform gathers flu related tweets from the entire world, with searching tag, words that are influenza synonyms and flu symptoms. The tags that being tracked are: Influenza, flu, chills, headache, sore throat, runny nose, sneezing, fever, dry cough. For every tweet extracted, additional amount of metadata is extracted too.
The system filters the Tweets automatically by removing tags and hashtags (including @, # symbols). Moreover, only geolocated tweets and tweets that their geolocation is extracted from the user’s profile location are being saved to the database. If enough location information is not available, tweets are automatically avoided. Tweets having less than 5 characters and those containing non-ASCII characters are excluded. Profile location is also automatically filtered from “suspicious” words (home, heaven etc), in consideration of avoiding false or non-exist location coordinates.
Since the above procedure is complete, tweets are visualized and updated every 20 minutes. New tweets are pinned in a worldwide map. In terms of city level, the Flutrack user can navigate through Google maps in order to examine where the tweets were exactly posted. Visuals are displayed on a styled Google map, free of unnecessary information and centred on mirroring the accurate position of tweets. The system displays new tweets and tweets that are until seven days old, with a deflection of 20 meters in real time scale coordinates.
Flutrack’s data are provided to every user via a simple JSON call.
In order to facilitate users, Flutrack’s API sorts and export tweets according to:
|http://api.flutrack.org/?s=feverANDcoughORfever||AND,OR operators supported|
|http://api.flutrack.org/?a=TRUE||only aggravated tweets to be returned, or not|
|http://api.flutrack.org/?time=7||Return results within last No. of days|
|http://api.flutrack.org/?limit=50||Absence of "limit" parameter will result a default of 10 results|
|JSON File||http://flutrack.org/results.json||Download the current visualized json file|
|Aggravation||"1", "true", "on" and "yes" for TRUE and "0", "false", "off", "no" for FALSE (case insensitive)||Absence of "a" parameter will lead to mixed results|
|Limit||"1", "true", "on" and "yes" for TRUE and "0", "false", "off", "no" for FALSE (case insensitive)||Absence of "a" parameter will lead to mixed results|
If you have any questions, project enquiries or something else on your mind,
feel free to send a mail and we will get back to you with our availability.