Computer Science M.Sc. students Shane Sims and Vanessa Putnam, and Professor Cristina Conati won the best paper award at the IJCAI 2019 Workshop on Humanizing AI held in Macau, China.
Their paper, titled "Detecting Confusion from Eye-Tracking Data with Recurrent Neural Networks" shows that training an RNN with raw eye-tracking data allows for a better classifier to be learned, as compared to previous methods that use engineered features. Training with raw data ensures that potentially valuable signals of confusion are available to be learned from, while using an RNN exploits the data's sequential nature. Their work is the first to use deep learning to detect an affective state from eye-tracking data.