Optimal Knowledge Component Extracting Model for Knowledge-Concept Graph Completion in Education

This is the official implementations of OKCE: Optimal Knowledge Component Extracting Model for Knowledge-Concept Graph Completion in Education(IEEE Access, 2023). For technical details, please refer to:

Optimal Knowledge Component Extracting Model for Knowledge-Concept Graph Completion in Education [DOI]
Hyunhee Choi, Hayun Lee, Minjeong Lee

(1) Setup

This code has been tested with Python 3.8, Pytorch 1.12, R 4.2.2, CUDA 11.2 and cuDNN 8.0 on Ubuntu 20.04.

• Clone the repository

git clone https://github.com/LAIVDATA-EdTech-R-D/OKCE.git

• Setup dependencies

conda create -n OKCE python=3.8
conda activate OKCE
deb https://cloud.r-project.org/bin/linux/ubuntu focal-cran40/
sudo apt-get update
sudo apt-get install r-base
cd OKCE
pip install -r requirements.txt

(2) KDD dataset

KDD dataset can be found here. Download the files named "bridge_to_algebra_2008_2009.zip". Unzip the folder and move it to /data.

• data pre-processing

Rscript ./PGM/data_preprocessing.R

(3) Training

• Optimal KC(Knowledge Concept) extracting

# python merge_main_nonFS.py --data_path=../data/algebra_ratio3000.csv --n_epochs=150

• Optimal KC(Knowledge Concept) extracting with feature selection

python merge_main.py --data_path=../data/algebra_ratio3000.csv --n_epochs=150