Publications

Using Similarity Learning with SBERT to Optimize Teacher Report Embeddings for Academic Performance Prediction

Published in AIED 2023, 2023

We propose a model that uses similarity learning as an embedding-enhancing technique. Results outperform earlier research with an average accuracy of 73% for detecting strong performance.

Recommended citation: Fateen, M., & Mine, T. (2023). Using Similarity Learning with SBERT to Optimize Teacher Report Embeddings for Academic Performance Prediction. In International Conference on Artificial Intelligence in Education (pp. 720-726). Cham: Springer Nature Switzerland. https://link.springer.com/chapter/10.1007/978-3-031-36336-8_111

Extraction of Useful Observational Features from Teacher Reports for Student Performance Prediction

Published in International Conference on Artificial Intelligence in Education, 2022

In this study, we analyze the topics, and psychological features in teachers’ daily written reports and apply them to the student performance prediction model. Experimental results show the capability of this approach in contributing to the accuracy of performance prediction models. We presented the paper in July 2022, at the International Conference on Artificial Intelligence in Education online.

Recommended citation: Fateen, M., & Mine, T. (2022) Extraction of Useful Observational Features from Teacher Reports for Student Performance Prediction.

An Improved Model to Predict Student Performance using Teacher Observation Reports

Published in International Conference on Computers in Education, 2021

In this paper, we present a model that improves on our previous model that predicts students’ performance using teacher observation reports. In this model, we take a more statistical approach where we summarize each students’ performance throughout the year. We presented the paper in November 2021, at the International Conference on Computers in Education online.

Recommended citation: Fateen, M., Ueno, K., & Mine, T. (2021) An Improved Model to Predict Student Performance using Teacher Observation Reports.

Predicting Student Performance Using Teacher Observation Reports

Published in Educational Data Mining 2021, 2021

In this paper, we present a model that predicts students’ performance using teacher observation reports. We compare the performance of the model using TF-IDF vectors and BERT embeddings. This paper is the first to utilize teacher reports for performance prediction. We presented the paper in June 2021, at the Educational Data Mining Conference online.

Recommended citation: Fateen, M., & Mine, T. (2021). Predicting Student Performance Using Teacher Observation Reports. International Educational Data Mining Society. https://eric.ed.gov/?id=ED615587