Neural Cognitive Diagnosis for Intelligent Education Systems
Authors: Fei Wang, Qi Liu, Enhong Chen, Zhenya Huang, Yuying Chen, Yu Yin, Zai Huang, Shijin Wang6153-6161
AAAI 2020 | Conference PDF | Archive PDF | Plain Text | LLM Run Details
| Reproducibility Variable | Result | LLM Response |
|---|---|---|
| Research Type | Experimental | Extensive experimental results on real-world datasets show the effectiveness of Neural CD framework with both accuracy and interpretability. |
| Researcher Affiliation | Collaboration | 1Anhui Province Key Lab. of Big Data Analysis and Application, School of Computer Science and Technology, University of Science and Technology of China 2i FLYTEK Research |
| Pseudocode | No | The paper describes the model architecture and provides mathematical formulations, but does not include structured pseudocode or algorithm blocks. |
| Open Source Code | Yes | Our code of Neural CDM is available at https://github.com/bigdata-ustc/Neural CD. |
| Open Datasets | Yes | ASSIST (ASSISTments 2009-2010 skill builder ) is an open dataset collected by the ASSISTments online tutoring systems (Feng, Heffernan, and Koedinger 2009), which only provides student response logs and knowledge concepts. We choose the public corrected version that eliminates the duplicated data issue proposed by previous work (Xiong et al. 2016). https://sites.google.com/site/assistmentsdata/home/ assistment-2009-2010-data/skill-builder-data-2009-2010 |
| Dataset Splits | Yes | We perform a 80%/20% train/test split of each student s response log. All models are evaluated with 5-fold cross validation. |
| Hardware Specification | Yes | All models are implemented by Py Torch using Python, and all experiments are run on a Linux server with four 2.0GHz Intel Xeon E5-2620 CPUs and a Tesla K20m GPU. |
| Software Dependencies | No | The paper mentions that models are implemented by 'Py Torch using Python', but does not specify version numbers for either PyTorch or Python, nor does it list any other software dependencies with version numbers. |
| Experiment Setup | Yes | The dimensions of the full connection layers (Eq. (7) (9)) are 512, 256, 1 respectively, and Sigmoid is used as activation function for all of the layers. We set hyperparameters λ = 0.1 (Eq. (12)) and σ = 1 ( Eq. (13)). For k in top-k knowledge concepts selecting, we use the value that make the predicting network reach 0.85 recall. That is, in our experiment, k = 20. We initialize the parameters with Xavier initialization (Glorot and Bengio 2010)... The CNN architecture we use in Neural CDM+ contains 3 convolutional layers... The channels of convolutional layers are 400, 200, 100, and kernel sizes are set to 3, 4, 5 respectively. We adopt Re Lu activation function for convolution layers and Sigmoid for the output layer. Multi-label binary cross entropy is used as loss function for training the CNN. |