Modeling Skill Acquisition Over Time with Sequence and Topic Modeling

José González-Brenes
Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, PMLR 38:296-305, 2015.

Abstract

Online education provides data from students solving problems at different levels of proficiency over time. Unfortunately, methods that use these data for inferring student knowledge rely on costly domain expertise. We propose three novel data-driven methods that bridge sequence modeling with topic models to infer students’ time varying knowledge. These methods differ in complexity, interpretability, accuracy and human supervision. For example, our most interpretable method has similar classification accuracy to the models created by domain experts, but requires much less effort. On the other hand, the most accurate method is completely data-driven and improves predictions by up to 15% in AUC, an evaluation metric for classifiers.

Cite this Paper

BibTeX
@InProceedings{pmlr-v38-gonzalez-brenes15, title = {{Modeling Skill Acquisition Over Time with Sequence and Topic Modeling}}, author = {González-Brenes, José}, booktitle = {Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics}, pages = {296--305}, year = {2015}, editor = {Lebanon, Guy and Vishwanathan, S. V. N.}, volume = {38}, series = {Proceedings of Machine Learning Research}, address = {San Diego, California, USA}, month = {09--12 May}, publisher = {PMLR}, pdf = {http://proceedings.mlr.press/v38/gonzalez-brenes15.pdf}, url = {https://proceedings.mlr.press/v38/gonzalez-brenes15.html}, abstract = {Online education provides data from students solving problems at different levels of proficiency over time. Unfortunately, methods that use these data for inferring student knowledge rely on costly domain expertise. We propose three novel data-driven methods that bridge sequence modeling with topic models to infer students’ time varying knowledge. These methods differ in complexity, interpretability, accuracy and human supervision. For example, our most interpretable method has similar classification accuracy to the models created by domain experts, but requires much less effort. On the other hand, the most accurate method is completely data-driven and improves predictions by up to 15% in AUC, an evaluation metric for classifiers.} }
Endnote
%0 Conference Paper %T Modeling Skill Acquisition Over Time with Sequence and Topic Modeling %A José González-Brenes %B Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics %C Proceedings of Machine Learning Research %D 2015 %E Guy Lebanon %E S. V. N. Vishwanathan %F pmlr-v38-gonzalez-brenes15 %I PMLR %P 296--305 %U https://proceedings.mlr.press/v38/gonzalez-brenes15.html %V 38 %X Online education provides data from students solving problems at different levels of proficiency over time. Unfortunately, methods that use these data for inferring student knowledge rely on costly domain expertise. We propose three novel data-driven methods that bridge sequence modeling with topic models to infer students’ time varying knowledge. These methods differ in complexity, interpretability, accuracy and human supervision. For example, our most interpretable method has similar classification accuracy to the models created by domain experts, but requires much less effort. On the other hand, the most accurate method is completely data-driven and improves predictions by up to 15% in AUC, an evaluation metric for classifiers.
RIS
TY - CPAPER TI - Modeling Skill Acquisition Over Time with Sequence and Topic Modeling AU - José González-Brenes BT - Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics DA - 2015/02/21 ED - Guy Lebanon ED - S. V. N. Vishwanathan ID - pmlr-v38-gonzalez-brenes15 PB - PMLR DP - Proceedings of Machine Learning Research VL - 38 SP - 296 EP - 305 L1 - http://proceedings.mlr.press/v38/gonzalez-brenes15.pdf UR - https://proceedings.mlr.press/v38/gonzalez-brenes15.html AB - Online education provides data from students solving problems at different levels of proficiency over time. Unfortunately, methods that use these data for inferring student knowledge rely on costly domain expertise. We propose three novel data-driven methods that bridge sequence modeling with topic models to infer students’ time varying knowledge. These methods differ in complexity, interpretability, accuracy and human supervision. For example, our most interpretable method has similar classification accuracy to the models created by domain experts, but requires much less effort. On the other hand, the most accurate method is completely data-driven and improves predictions by up to 15% in AUC, an evaluation metric for classifiers. ER -
APA
González-Brenes, J.. (2015). Modeling Skill Acquisition Over Time with Sequence and Topic Modeling. Proceedings of the Eighteenth International Conference on Artificial Intelligence and Statistics, in Proceedings of Machine Learning Research 38:296-305 Available from https://proceedings.mlr.press/v38/gonzalez-brenes15.html.

Related Material