Activity
Time:
Graduate Defense: Sumeet Maan
Date:
10:00 am –
11:00 am
Zoom
Graduate Defense: Sumeet Maan
Monday, November 16, 2020
10 a.m.
Zoom: https://go.unl.edu/kuzv
Meeting ID: 966 4433 4133
Title: “Representational Learning Approach for Predicting Developer Expertise Using Eye Movements”
Abstract: The thesis introduces a unique way of feature engineering from eye-tracking data and predicts developer expertise using that data. The eye gaze data of developers was collected while they were trying to solve a realistic bug. The data differ in complexity and feature-length. The thesis provides a comparative analysis of representational learning and non-representational learning. The results are obtained from an extensive set of experiments that show that representational learning works well on the dataset. Furthermore, a state-of-the-art method for feature engineering is proposed to extract features from the eye-tracking data. The average accuracy on all the tasks is 93.4\% with 78.8\% recall and 81.6\% F1 score.
Monday, November 16, 2020
10 a.m.
Zoom: https://go.unl.edu/kuzv
Meeting ID: 966 4433 4133
Title: “Representational Learning Approach for Predicting Developer Expertise Using Eye Movements”
Abstract: The thesis introduces a unique way of feature engineering from eye-tracking data and predicts developer expertise using that data. The eye gaze data of developers was collected while they were trying to solve a realistic bug. The data differ in complexity and feature-length. The thesis provides a comparative analysis of representational learning and non-representational learning. The results are obtained from an extensive set of experiments that show that representational learning works well on the dataset. Furthermore, a state-of-the-art method for feature engineering is proposed to extract features from the eye-tracking data. The average accuracy on all the tasks is 93.4\% with 78.8\% recall and 81.6\% F1 score.