The influence of social media has given rise to a new form of digital literacy. While many youth are digitally literate in a traditional sense, they lack an understanding of how the information they share online can be used to identify them. Through a generous grant from the Canadian Internet Registration Authority’s (CIRA) Community Investment Program we will develop an interactive educational resource that demonstrates how publicly shared textual content can lead to the identification of individuals through exposing their locations.
Every word in any language is situated in a geographic context. In the Platial Analysis Lab, we build probabilistic models for identifying the “geo-indicativeness” of text using novel machine learning techniques applied to big, public, geosocial datasets. To use an overly simple example, content containing the word “taco” is slightly more likely to be written by someone in Baldwin Village, Toronto, than any other place in Canada. Our extensive research in this area demonstrates that multiplying word spatial probabilities, even for short sentences, increases the ability to locate someone.
Through the proposed platform, students will enter text into a simple interface (e.g, a social media post) and the LocatEd platform will inform them, numerically and cartographically, how much location information they are exposing given their choice of terms.