Week 1 — Eye Tracking and Prior Knowledge

by Alper Özöner and Ali Utku Aydın

This week, we finally chose our topic which is to create a ML model that predicts prior knowledge of a certain topic by analyzing people’s reading patterns as they read a short text about a given topic (such as history, politics, or science) all the while tracking their pupils with a 4000$-cheap IR sensor called the Eye Tribe, enshrined to us by ever so generous Kayhan Hoca, for our part-time research work.

On Friday, after a quick meeting with Mr. Imre, we managed to record the tracking data directly from the sensor to my laptop and logged the data in a .txt file by using the default API.

Our second day of work was more tiring than anticipated as we struggled through the ever-alien formatting of the aforementioned .txt file. Thanks to Utku’s struggles, we managed to read a measly total of 2 columns out of 36. But I guess it’s a start, nonetheless.