Research on the Difficulty Prediction for Questions Based on Learners’ Response

DOI

10.2991/978-94-6463-502-7_46How to use a DOI?

Keywords

difficulty prediction; deep learning; learners’ response; intelligent education

Abstract

The question difficulty assessment is an important research direction in educational data mining. The traditional difficulty assessment is often completed manually, which is time-consuming and subjective. Existing difficulty prediction models are usually limited to the final score or only for statistical analysis of question text, without combining learners’ responses and question context or detailed information. They often cannot effectively reflect the difference between the cognition of the question and the difficulty of the question, unable to meet the requirements of the instructional practice. Therefore, this paper aims to propose a question difficulty prediction model based on learners’ responses and combine natural language processing technology to realize the automatic prediction of question difficulty. Specifically, the paper first based on the BERT training model, extracts the question information embedded vector, combined with convolutional neural network and long and short-term memory network, the fusion of learners’ response (including score, response time, submit time, etc.), establish the correlation between the question text information and the question difficulty, construct the difficulty prediction of questions model based on learners’ response, and achieve accurate question difficulty prediction.

Copyright

© 2024 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  - Jiaqi Long
AU  - Hui Zhao
AU  - Jie Pu
AU  - Yifan Liu
AU  - Binghan Ju
AU  - Suojuan Zhang
PY  - 2024
DA  - 2024/08/31
TI  - Research on the Difficulty Prediction for Questions Based on Learners’ Response
BT  - Proceedings of the 2024 5th International Conference on Education, Knowledge and Information Management (ICEKIM 2024）
PB  - Atlantis Press
SP  - 440
EP  - 453
SN  - 2589-4900
UR  - https://doi.org/10.2991/978-94-6463-502-7_46
DO  - 10.2991/978-94-6463-502-7_46
ID  - Long2024
ER  -