Monitored Analysis for eXtreme eXpressions
An automated web-scraper which analyzes digital expression (i.e. Twitter posts) for linguistic patterns which have been statistically related to extreme states of mind, such as mental illness or risk of harming one’s self and/or others.
Input: Twitter account
Output: Risk assessment of given account
Algorithm: Monitors for and identifies certain linguistic factors which may be cause for concern
Activates an alert if: a certain amount of risk factors are identified, and/or many risk factors are used in combination with each other
Future plans: Generalize for use with platforms other than Twitter (Facebook, Reddit, Instagram) Refine the analysis algorithm Base the algorithm purely on hard research
Devpost: https://devpost.com/software/maxx