Automatic Dialect Extraction through supervised learning using IBM Watson

This project constituted a major portion of my IBM Watson class. Where I led a team of colleagues to create this system throughout the 2016 Spring semester at UT Dallas

The purpose of the project is train watson to identify the nationality of the author of a given Spanish text input. One scenario where this project can prove helpful could be in helping authorities identify the nationality of a text conversation, or an email thread simply by the way they express themselves in text and narrowing the suspect’s possible identity down to a nationality level. Watson will be trained to identify 5 different Spanish speaking countries.

A quick demo of it is [here](

Originally published at on December 10, 2016.