Research on Student Academic Performance Prediction and Improvement Strategies Based on LSTM

DOI

10.2991/978-94-6239-691-3_21How to use a DOI?

Keywords

Student; Long Short-Term Memory networks; Academic performance prediction; Improvement strategies

Abstract

This study proposes an academic early warning and intervention framework based on LSTM and behavioral features. The framework first leverages study duration, attendance rates, sleep hours, and historical grades to construct a multi-dimensional dataset and trains an LSTM prediction model, which achieves an RMSE of 3.1747 on the test set, demonstrating high predictive precision. Furthermore, the study transforms the trained model into an interactive simulation environment to conduct feature intervention simulations targeting three student personas: “High Effort, Low Efficiency,” “Insufficient Investment,” and “Weak Foundation.” Experiments demonstrate that tailored improvement strategies can significantly enhance predicted performance, yielding a maximum score increase of 8.9 points, thereby achieving a transition from static grade prediction to dynamic personalized strategy generation.

Copyright

© 2026 The Author(s)

Open Access

Open Access This chapter is licensed under the terms of the Creative Commons Attribution-NonCommercial 4.0 International License (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommercial use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license and indicate if changes were made.

Download article (PDF)

TY  - CONF
AU  - Jianwei Huang
AU  - Yanyu Huang
PY  - 2026
DA  - 2026/05/31
TI  - Research on Student Academic Performance Prediction and Improvement Strategies Based on LSTM
BT  - Proceedings of the 2026 5th International Conference on Educational Innovation and Multimedia Technology (EIMT 2026)
PB  - Atlantis Press
SP  - 199
EP  - 205
SN  - 2667-128X
UR  - https://doi.org/10.2991/978-94-6239-691-3_21
DO  - 10.2991/978-94-6239-691-3_21
ID  - Huang2026
ER  -