Filters








9 Hits in 4.7 sec

SAINT+: Integrating Temporal Features for EdNet Correctness Prediction [article]

Dongmin Shin, Yugeun Shim, Hangyeol Yu, Seewoo Lee, Byungsoo Kim, Youngduck Choi
2020 arXiv   pre-print
Moreover, SAINT+ incorporates two temporal feature embeddings into the response embeddings: elapsed time, the time taken for a student to answer, and lag time, the time interval between adjacent learning  ...  We empirically evaluate the effectiveness of SAINT+ on EdNet, the largest publicly available benchmark dataset in the education domain.  ...  Also, SAINT+ improves SAINT with 1.03% and 1.25% gain in ACC and AUC, respectively, demonstrating the effectiveness of integrating the temporal features for knowledge tracing.  ... 
arXiv:2010.12042v1 fatcat:g22tl3cft5fo7pmdldcqf6g5ka

Assessing the Performance of Online Students – New Data, New Approaches, Improved Accuracy [article]

Robin Schmucker, Jingbo Wang, Shijia Hu, Tom M. Mitchell
2022 arXiv   pre-print
to create a group prediction of the SP.  ...  Second, we take advantage of features of the student history that go beyond question-response pairs (e.g., features such as which video segments the student watched, or skipped) as well as information  ...  INTEGRATING FEATURES INTO LOGISTIC REGRESSION We have seen how individual features can be integrated into the Best-LR model to increase prediction quality.  ... 
arXiv:2109.01753v2 fatcat:jcazup62nza5jltkjoqdx2dzj4

Assessing the Performance of Online Students - New Data, New Approaches, Improved Accuracy

Robin Schmucker, Jingbo Wang, Shijia Hu, Tom M. Mitchell
2022 Zenodo  
models to create a group prediction of the student performance.  ...  This student performance modeling problem is a critical step for building adaptive online teaching systems.  ...  DAS3H incorporates a temporal aspect into its predictions by computing count features for different time windows.  ... 
doi:10.5281/zenodo.6450190 fatcat:ecqrsqhskbf3fbvjfacykwzkza

Code-DKT: A Code-based Knowledge Tracing Model for Programming Tasks

Yang Shi, Min Chi, Tiffany Barnes, Thomas Price, Antonija Mitrovic, Nigel Bosch
2022 Zenodo  
Finally, we analyze problem-specific performance and a set of case studies for one assignment to demonstrate when and how code features improve the Code-DKT's predictions.  ...  Knowledge tracing (KT) models are a popular approach for predicting students' future performance at practice problems using their prior attempts.  ...  For perspective, this improvement is comparable to SAINT+'s improvement over DKT on EdNet (+2.76%) [13] , or SAKT's improvement on various datasets (+3.8%) .  ... 
doi:10.5281/zenodo.6853105 fatcat:jdtkbyiyprf4hkld64xcblp77a

Code-DKT: A Code-based Knowledge Tracing Model for Programming Tasks [article]

Yang Shi, Min Chi, Tiffany Barnes, Thomas Price
2022 arXiv   pre-print
Finally, we analyze problem-specific performance through a set of case studies for one assignment to demonstrate when and how code features improve Code-DKT's predictions.  ...  Knowledge tracing (KT) models are a popular approach for predicting students' future performance at practice problems using their prior attempts.  ...  For perspective, this improvement is comparable to SAINT+'s improvement over DKT on EdNet (+2.76%) [13] , or SAKT's improvement on various datasets (+3.8%) .  ... 
arXiv:2206.03545v1 fatcat:ld2b4w4gdvaczcone2d4o4ufhi

A Survey of Knowledge Tracing [article]

Qi Liu, Shuanghong Shen, Zhenya Huang, Enhong Chen, Yonghe Zheng
2021 arXiv   pre-print
Finally, we provide some potential directions for future research in this fast-growing field.  ...  Both SAINT and SAINT+ have achieved superior performance relative to SAKT on the EdNet dataset [58] , one of the largest publicly available datasets for educational data mining.  ...  Subsequently, the SAINT+ model [57] is proposed to incorporate two temporal features into SAINT: namely, the answering time for each exercise and the interval time between two continuous learning interactions  ... 
arXiv:2105.15106v2 fatcat:723wl2krqzd3ziboc2vmhdu23q

Empirical Evaluation of Deep Learning Models for Knowledge Tracing: Of Hyperparameters and Metrics on Performance and Replicability [article]

Sami Sarsa, Juho Leinonen, Arto Hellas
2022 arXiv   pre-print
value may outperform DLKT models, especially in terms of accuracy -- highlighting importance of selecting proper baselines for comparison; Disambiguation of properties that affect performance in DLKT  ...  contributions of this work are: Evidence that DLKT models generally outperform more traditional models, but not necessarily by much and not always; Evidence that even simple baselines with little to no predictive  ...  We are grateful for the grant by the Media Industry Research Foundation of Finland which partially funded this work. We thank the reviewers for their valuable comments that helped improved this  ... 
arXiv:2112.15072v4 fatcat:i6pkahldsnbmxn45zpvtyk5zf4

Abstracts From the 24th Annual Health Care Systems Research Network Conference, April 11–13, 2018, Minneapolis, Minnesota

2018 Journal of Patient-Centered Research and Reviews  
This model for ODA encourages a coherent approach focused on support for the self-correcting mechanism in science and optimization of inferential integrity.  ...  Numerous threats militate against validity, reproducibility, integrity of inference, and the self-correcting mechanism of the scientific method.  ...  Methods: We implemented the SuperShelf model in two Minneapolis/Saint Paul metro area food shelves.  ... 
doi:10.17294/2330-0698.1630 fatcat:ucdf7vazvjdwnd3zrkpob46fp4

NF-kappaB and apoptosis

1996 Science  
The remainder of the day will feature a career fair for employers and job-seekers.  ...  Experience with structure-property relationships and solubility predictions is a plus. To be considered for this position, please reference Dept. TM-PHB-JW.  ...  Visit our web site at http://www.resgen.com or call and ask for Kay Swanson or Jim Hudson for more information.  ... 
doi:10.1126/science.274.5288.697e fatcat:zlobquedlnbvjmkqhrqofnmany