Cognitive Modelling for Predicting Examinee Performance

Authors: Runze Wu, Qi Liu, Yuping Liu, Enhong Chen, Yu Su, Zhigang Chen, Guoping Hu

IJCAI 2015 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on three realworld datasets prove that Fuzzy CDF can predict examinee performance more effectively, and the output of Fuzzy CDF is also interpretative.
Researcher Affiliation Collaboration Runze Wu1, Qi Liu1, , Yuping Liu1, Enhong Chen1, Yu Su2, Zhigang Chen2, Guoping Hu2 1School of Computer Science and Technology, University of Science and Technology of China {wrz179,liuyup}@mail.ustc.edu.cn, {qiliuql,cheneh}@ustc.edu.cn 2Anhui USTC i FLYTEK Co., Ltd., China {yusu,zgchen,gphu}@iflytek.com
Pseudocode Yes Algorithm 1 Sampling algorithm for Fuzzy CDF.
Open Source Code No The paper states: "The two private datasets we use have been publicly available on http://staff.ustc.edu.cn/%7Eqiliuql/data/math2015.rar." This provides access to datasets, not the source code for the methodology described in the paper. There is no explicit statement or link indicating that the authors' implementation code is open-source or publicly available.
Open Datasets Yes The public dataset in our experiment comprises of scores of middle school students on fraction subtraction objective problems [Tatsuoka, 1984; De Carlo, 2010]. The two private datasets2 are collected from two final mathematical exams for high school students including both objective and subjective problems. We denote the three datasets as Frc Sub, Math1 and Math2. ... 2The two private datasets we use have been publicly available on http://staff.ustc.edu.cn/%7Eqiliuql/data/math2015.rar.
Dataset Splits No To observe how the methods behave at different sparsity levels, we construct different sizes of training sets, with 20%, 40%, 60% and 80% of score data of each examinee, and the rest for testing, respectively.
Hardware Specification Yes Both our Fuzzy CDF and other baseline approaches are implemented by using Matlab on a Core i5 3.1Ghz machine with Windows 7 and 4 GB memory.
Software Dependencies No Both our Fuzzy CDF and other baseline approaches are implemented by using Matlab on a Core i5 3.1Ghz machine with Windows 7 and 4 GB memory.
Experiment Setup Yes For the prior distributions of parameters in Fuzzy CDF, we set the hyperparameters as follows: µθ = 0, σθ = 1; µa = 0, σa = 1; µb = 0, σb = 1; vs = 1, ws = 2, mins = 0, maxs = 0.6; vg = 1, wg = 2, ming = 0, maxg = 0.6; xσ = 4, yσ = 6. In this experiments, we set the number of iterations of Algorithm 1 to 5,000 and estimate the parameters based on the last 2,500 samples to guarantee the convergency of the Markov chain.