IKT

https://arxiv.org/pdf/2112.11209.pdf. In Proceedings of the 36th AAAI Conference on Artificial Intelligence: EAAI.

Data format

First line : the number of skills a student attempted with student ID. Second line : the skill id sequence. Third line : the question id sequence. Forth line : the response sequence.

    15,1
    1,1,1,1,7,7,9,10,10,10,10,11,11,45,54
    5001,5023,5044,5066,2014,2058,6017,20004,60001,1200,1201,1311,2014,2410,2001
    0,1,1,1,1,1,0,0,1,1,1,1,1,0,0

Feature Engineering

1) Run FeatureEngineering.py

produce train_data.csv and test_data.csv

Transform CSV to Arff (easy way to transform)

1)Copy the follwing to CSV header

@relation ASS2009
@attribute skill_ID numeric
@attribute skill_mastery numeric
@attribute ability_profile numeric
@attribute problem_difficulty numeric
@attribute correctness {1,0}
@data

2) change both train and test files extension from .csv to .arff

WEKA

1) Install WEKA

Run TANB classifer with WEKA

After getting the data set with features and class variables. Run TAN classifier under train and test setting

1) preprocess tag-> open file: training data 2) classify tag->choose: weka>classifier>bayes>BayesNet->searchAlgorithm: TAN 3) supply test set with test data 4) start to run

Train and Test Datasets used in this paper can be found at the following link.

https://drive.google.com/drive/folders/1Wuilcb_ash1r5MT3tgMc78PDPUh0n3uT?usp=sharing