Machine Learning-Based Prediction of the Course Assessment

Weimin Geng, Qiuling Li, Dian Zhang

2024

Abstract

In order to keep track of the students' learning status and make early warning, the model of predicting the final course assessment was proposed based on machine learning. Take the course of water supply and drainage engineering cost as an example, the students’ related information and the historical assessment data (such as the teaching activities and the stage assessment scores etc.) collection and cleaning were carried out firstly. Then the features were filtered out by Light Gradient Boosting Machine (LightGBM), and the prediction model of the final score was built on the basis of Least Absolute Shrinkage and Selection Operator (Lasso) regression. Phases 1 and 2 forecast were completed and the error statistics were analysed. The predicted results at different stages of the semester help the teachers and students get the learning situation and take timely adjustment measures.

Paper Citation

in Harvard Style

Geng W., Li Q. and Zhang D. (2024). Machine Learning-Based Prediction of the Course Assessment. In Proceedings of the 7th International Conference on Environmental Science and Civil Engineering - Volume 1: ICESCE; ISBN 978-989-758-764-1, SciTePress, pages 168-172. DOI: 10.5220/0013593400004671

in Bibtex Style

@conference{icesce24,
author={Weimin Geng and Qiuling Li and Dian Zhang},
title={Machine Learning-Based Prediction of the Course Assessment},
booktitle={Proceedings of the 7th International Conference on Environmental Science and Civil Engineering - Volume 1: ICESCE},
year={2024},
pages={168-172},
publisher={SciTePress},
organization={INSTICC},
doi={10.5220/0013593400004671},
isbn={978-989-758-764-1},
}

in EndNote Style

TY - CONF

JO - Proceedings of the 7th International Conference on Environmental Science and Civil Engineering - Volume 1: ICESCE
TI - Machine Learning-Based Prediction of the Course Assessment
SN - 978-989-758-764-1
AU - Geng W.
AU - Li Q.
AU - Zhang D.
PY - 2024
SP - 168
EP - 172
DO - 10.5220/0013593400004671
PB - SciTePress