Twitter as a Sensor Network

You’d like to think there is such a thing as synchronicity. Really, you would. The universe could then be counted upon for providing the necessary infrastructural support for you to fulfill your destiny. (Try to say that out loud without rasping like Darth Vader.) Or, alternatively, it could be readily blamed for any failure to achieve said destiny. (Or, on yet another hand that we do not as humans possess, the universe might respond that your failure is really one to recognize your true destiny, which is almost always a little (a lot) less fulfilling than the ones you glimpse in novels or movies.) But, really, you should put down that book, or turn off whatever screen you are watching, and get back to working on your destiny.

Or maybe it isn’t synchronicity but simply the fact that you are now aware of some particular thing. Two weeks ago I attended a small conference sponsored by the Louisiana Board of Regents and ended up in a fascinating conversation with someone. Posting someone’s name without their permission is against my policy, and so I will simply note that my own interest in using digital textual analysis tools to non-literary discourses meshed well with that person’s interest in seeking ways to sift through data generated by humans in order to determine the nature of an event.

With that conversation, and a follow-up lunch, in my head, I came across this [post on I Programmer][post] by Mike James that details the efforts of Rice University researchers to use Twitter as a “multimodal sensor network”:

> Can people be used as a “sensor net” to detect when important events happen? When it comes to sporting events it seems that all you have to do is look to the Twitter frequency.
The big thing for the near future is the Internet of Things but a group of researchers at Rice University (Houston) think that we are being a bit narrow in the meaning of the term “thing”. To quote from their paper: “The global human population can be regarded as geographically distributed, multimodal sensors.”

The post gives their method in a nutshell, and, thankfully, also has a link to a fuller account: