Exercise-Enhanced Sequential Modeling for Student Performance Prediction
Authors: Yu Su, Qingwen Liu, Qi Liu, Zhenya Huang, Yu Yin, Enhong Chen, Chris Ding, Si Wei, Guoping Hu
AAAI 2018 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experiments on large-scale real-world data clearly demonstrate the effectiveness of EERNN framework. Moreover, by incorporating the exercise correlations, EERNN can well deal with the cold start problems from both student and exercise perspectives. |
| Researcher Affiliation | Collaboration | Yu Su, Qingwen Liu, Qi Liu, Zhenya Huang, Yu Yin, Enhong Chen, Chris Ding, Si Wei, Guoping Hu School of Computer Science and Technology, Anhui University Anhui Province Key Lab. of Big Data Analysis and Application, University of Science and Technology of China i FLYTEK Research, CSE Department, University of Texas at Arlington {yusu, qwliu}@iflytek.com, qiliuql@ustc.edu.cn, {huangzhy, yxonic}@mail.ustc.edu.cn cheneh@ustc.edu.cn, chqding@uta.edu, {siwei, gphu}@iflytek.com |
| Pseudocode | No | The paper describes models using mathematical equations and figures, but no structured pseudocode or algorithm blocks are provided. |
| Open Source Code | No | The paper does not provide any statements about releasing source code, nor does it include links to a code repository. |
| Open Datasets | No | The experimental dataset supplied by i FLYTEK Co., Ltd. is collected from Zhixue1, a widely-used online learning system, which provides senior high school students with a large exercise resources for exercising. The dataset is referred to as "mathematical dataset" from this system, but no public access link or citation to a public repository is provided. |
| Dataset Splits | No | The paper states: 'For each student s sequential exercising record, we use the beginning 60%, 70%, 80%, 90% exercises as training sets, and the remains are as testing sets, respectively.' While it mentions training and testing splits, it does not explicitly provide specific percentages or counts for a separate validation dataset split. |
| Hardware Specification | Yes | All models are implemented by Py Torch (Paszke and Chintala ) using Python on a Linux server with four 2.0GHz Intel Xeon E5-2620 CPUs and a Tesla K20m GPU. |
| Software Dependencies | No | The paper mentions 'Py Torch' and 'Python' and 'word2vec tool' but does not provide specific version numbers for these software dependencies (e.g., PyTorch 1.x or Python 3.x). |
| Experiment Setup | Yes | We set the dimension dv of hidden states in Exercise Embedding as 100, dh of hidden states in Student Embedding as 100, and dy of overall presentation vectors in prediction stage as 50, respectively. Besides, we set mini batches as 32 for training and also use dropout (with probability 0.1) to prevent overfitting. |