Learner Modeling - Revision history

Yuh43: /* Hierarchical Knowledge Structure Modeling */

2016-12-04T16:31:02Z

‎Hierarchical Knowledge Structure Modeling

Yuh43 at 16:28, 4 December 2016

2016-12-04T16:28:56Z

Yuh43 at 16:28, 4 December 2016

2016-12-04T16:28:32Z

Yuh43 at 16:25, 4 December 2016

2016-12-04T16:25:17Z

Yuh43: /* Dynamic Knowledge Modeling in Textbook-Based Learning */

2016-12-04T16:24:51Z

‎Dynamic Knowledge Modeling in Textbook-Based Learning

Yuh43 at 16:24, 4 December 2016

2016-12-04T16:24:29Z

Yuh43: /* Introduction */

2016-04-05T04:22:46Z

‎Introduction

Yuh43 at 04:22, 5 April 2016

2016-04-05T04:22:01Z

Yuh43 at 04:21, 5 April 2016

2016-04-05T04:21:17Z

Yuh43: /* Asymptotic Assessment of User Knowledge */

2016-04-05T04:12:25Z

‎Asymptotic Assessment of User Knowledge

@@ Line 55: / Line 55: @@
 * Gonzalez-Brenes, J. P., Huang, Y. Using Data from Real and Simulated Learners to Evaluate Adaptive Tutoring Systems. In: 2nd AIED Workshop on Simulated Learners at the 17th Intl. Conf. on Artificial Intelligence in Education (AIED 2015), Madrid, Spain, 2015. ([http://ceur-ws.org/Vol-1432/sl_pap4.pdf paper])
-== Hierarchical Knowledge Structure Modeling ==
+== Hierarchical Integration Skill Modeling ==
 === Introduction ===
 This work explores the problem of modeling student knowledge in complex learning activities where multiple skills are required at the same time, such as in the programming domain. In such cases, it is not clear how the evidence of student performance translates to individual skills. As a result, traditional approaches to knowledge modeling, such as Knowledge Tracing (KT), which traces students’ knowledge of each decomposed individual skill, might fall short. We argue that skill combinations might carry extra specific knowledge, and mastery should be asserted only when a student can fluently apply skills in combination with other skills in different contexts. We propose a data-driven framework to model skill combination patterns for tracing students’ deeper knowledge. We automatically identify significant skill combinations from data and construct a conjunctive knowledge model with a hierarchical skill representation based on a Bayesian Network. We also propose a novel evaluation framework primarily focuses on the knowledge inference quality, since we argue that traditional prediction metrics no longer suffice to differentiate between shallow and deep knowledge modeling. Our experiments on datasets collected from two programming learning systems show that proposed model significantly increases mastery inference accuracy and tends to more reasonably distribute students’ efforts comparing with traditional KT models and its nonhierarchical counterparts. Our work serves as a first step towards building skill application context sensitive model for modeling students’ deep, robust learning.
@@ Line 61: / Line 61: @@
 === Publications ===
 * Huang, Y., Guerra-Hollstein, J., and Brusilovsky, P. Modeling Skill Combination Patterns for Deeper Knowledge Tracing. In: The 6th Intl. Workshop on Personalization Approaches in Learning Environments (PALE 2016) in the 24th Conf. on User Modeling, Adaptation and Personalization (UMAP 2016). ([http://ceur-ws.org/Vol-1618/PALE4.pdf paper])
 == Content Model Reduction for Better Learner Models ==

@@ Line 34: / Line 34: @@
 * Huang, Yun, Michael Yudelson, Shuguang Han, Daqing He, and Peter Brusilovsky. "A Framework for Dynamic Knowledge Modeling in Textbook-Based Learning." In Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization, pp. 141-150. ACM, 2016. ([http://columbus.exp.sis.pitt.edu/yunhuang/papers/UMAP16.pdf paper]) ([http://www.slideshare.net/huangyun/umap16-a-framework-for-dynamic-knowledge-modeling-in-textbookbased-learning presentation])
-== Multifaceted Evaluation of Learner Models for Practitioners ==
+== Multifaceted Evaluation of Learner Models ==
 We have proposed two state-of-the-art frameworks for evaluating student models in a data-driven manner for practitioners: the Polygon framework, and the Learner Effort-Outcomes Paradigm. We also explored challenges of using observational data to answer research question such as determine the importance of example usage.

@@ Line 33: / Line 33: @@
 === Publications ===
 * Huang, Yun, Michael Yudelson, Shuguang Han, Daqing He, and Peter Brusilovsky. "A Framework for Dynamic Knowledge Modeling in Textbook-Based Learning." In Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization, pp. 141-150. ACM, 2016. ([http://columbus.exp.sis.pitt.edu/yunhuang/papers/UMAP16.pdf paper]) ([http://www.slideshare.net/huangyun/umap16-a-framework-for-dynamic-knowledge-modeling-in-textbookbased-learning presentation])
 == Multifaceted Evaluation of Learner Models for Practitioners ==
@@ Line 61: / Line 54: @@
 * Gonzalez-Brenes, J. P., Huang, Y. The Leopard Framework: Towards understanding educational technology interventions with a Pareto Efficiency Perspective. In: The ICML 2015 Workshop on Machine Learning for Education (ICML 2015), Lille, France, 2015. ([https://dsp.rice.edu/sites/dsp.rice.edu/files/leopard_evaluation(1).pdf paper])
 * Gonzalez-Brenes, J. P., Huang, Y. Using Data from Real and Simulated Learners to Evaluate Adaptive Tutoring Systems. In: 2nd AIED Workshop on Simulated Learners at the 17th Intl. Conf. on Artificial Intelligence in Education (AIED 2015), Madrid, Spain, 2015. ([http://ceur-ws.org/Vol-1432/sl_pap4.pdf paper])
 == Graph Analysis of Learner Models ==

@@ Line 29: / Line 29: @@
 == Dynamic Knowledge Modeling in Textbook-Based Learning ==
 === Introduction ===
-Various e-learning systems that provide electronic textbooks are gathering data on large numbers of student reading interactions. This data can potentially be used to model student knowledge acquisition. However, reading activity is often overlooked in canonical student modeling. Prior studies modeling learning from reading either estimate student knowledge at the end of all reading activities, or use quiz performance data with expert-crafted knowledge components (KCs). In this work, we demonstrate that the dynamic modeling of student knowledge is feasible and that automatic text analysis can be applied to save expert effort. We propose a data-driven approach for dynamic student modeling in textbook-based learning. We formulate the problem of modeling learning from reading as a reading- time prediction problem, reconstruct existing popular student models (such as Knowledge Tracing) and explore two automatic text analysis approaches (bag-of-words-based and latent semantic-based) to build the KC model. We evaluate the proposed framework using a dataset collected from a Human-Computer Interaction course. Results show that our approach for reading modeling is plausible; the pro- posed Knowledge Tracing-based student model reliably outperforms baselines and the latent semantic-based approach can be a promising way to construct a KC model. Serving as the first step to model dynamic knowledge in textbook- based learning, our framework can be applied to a broader context of open-corpus personalized learning.  [[Dynamic Knowledge Modeling in Textbook-Based Learning|More about dynamic knowledge modeling in textbook-based learning]]
+Various e-learning systems that provide electronic textbooks are gathering data on large numbers of student reading interactions. This data can potentially be used to model student knowledge acquisition. However, reading activity is often overlooked in canonical student modeling. Prior studies modeling learning from reading either estimate student knowledge at the end of all reading activities, or use quiz performance data with expert-crafted knowledge components (KCs). In this work, we demonstrate that the dynamic modeling of student knowledge is feasible and that automatic text analysis can be applied to save expert effort. We propose a data-driven approach for dynamic student modeling in textbook-based learning. We formulate the problem of modeling learning from reading as a reading- time prediction problem, reconstruct existing popular student models (such as Knowledge Tracing) and explore two automatic text analysis approaches (bag-of-words-based and latent semantic-based) to build the KC model. We evaluate the proposed framework using a dataset collected from a Human-Computer Interaction course. Results show that our approach for reading modeling is plausible; the pro- posed Knowledge Tracing-based student model reliably outperforms baselines and the latent semantic-based approach can be a promising way to construct a KC model. Serving as the first step to model dynamic knowledge in textbook- based learning, our framework can be applied to a broader context of open-corpus personalized learning.   [[Dynamic Knowledge Modeling in Textbook-Based Learning|==> more]]
 === Publications ===

@@ Line 28: / Line 28: @@
 == Dynamic Knowledge Modeling in Textbook-Based Learning ==
 === Introduction ===
-Various e-learning systems that provide electronic textbooks are gathering data on large numbers of student reading interactions. This data can potentially be used to model student knowledge acquisition. However, reading activity is often overlooked in canonical student modeling. Prior studies modeling learning from reading either estimate student knowledge at the end of all reading activities, or use quiz performance data with expert-crafted knowledge components (KCs). In this work, we demonstrate that the dynamic modeling of student knowledge is feasible and that automatic text analysis can be applied to save expert effort. We propose a data-driven approach for dynamic student modeling in textbook-based learning. We formulate the problem of modeling learning from reading as a reading- time prediction problem, reconstruct existing popular student models (such as Knowledge Tracing) and explore two automatic text analysis approaches (bag-of-words-based and latent semantic-based) to build the KC model. We evaluate the proposed framework using a dataset collected from a Human-Computer Interaction course. Results show that our approach for reading modeling is plausible; the pro- posed Knowledge Tracing-based student model reliably outperforms baselines and the latent semantic-based approach can be a promising way to construct a KC model. Serving as the first step to model dynamic knowledge in textbook- based learning, our framework can be applied to a broader context of open-corpus personalized learning.
+Various e-learning systems that provide electronic textbooks are gathering data on large numbers of student reading interactions. This data can potentially be used to model student knowledge acquisition. However, reading activity is often overlooked in canonical student modeling. Prior studies modeling learning from reading either estimate student knowledge at the end of all reading activities, or use quiz performance data with expert-crafted knowledge components (KCs). In this work, we demonstrate that the dynamic modeling of student knowledge is feasible and that automatic text analysis can be applied to save expert effort. We propose a data-driven approach for dynamic student modeling in textbook-based learning. We formulate the problem of modeling learning from reading as a reading- time prediction problem, reconstruct existing popular student models (such as Knowledge Tracing) and explore two automatic text analysis approaches (bag-of-words-based and latent semantic-based) to build the KC model. We evaluate the proposed framework using a dataset collected from a Human-Computer Interaction course. Results show that our approach for reading modeling is plausible; the pro- posed Knowledge Tracing-based student model reliably outperforms baselines and the latent semantic-based approach can be a promising way to construct a KC model. Serving as the first step to model dynamic knowledge in textbook- based learning, our framework can be applied to a broader context of open-corpus personalized learning.  [[Dynamic Knowledge Modeling in Textbook-Based Learning|More about dynamic knowledge modeling in textbook-based learning]]
 === Publications ===
 * Huang, Yun, Michael Yudelson, Shuguang Han, Daqing He, and Peter Brusilovsky. "A Framework for Dynamic Knowledge Modeling in Textbook-Based Learning." In Proceedings of the 2016 Conference on User Modeling Adaptation and Personalization, pp. 141-150. ACM, 2016. ([http://columbus.exp.sis.pitt.edu/yunhuang/papers/UMAP16.pdf paper]) ([http://www.slideshare.net/huangyun/umap16-a-framework-for-dynamic-knowledge-modeling-in-textbookbased-learning presentation])
 == Content Model Reduction for Better Learner Models ==

@@ Line 26: / Line 26: @@
 * Huang, Y., González-Brenes, J. P., and Brusilovsky, P. (2015) The FAST toolkit for Unsupervised Learning of HMMs with Features. In: The Machine Learning Open Source Software workshop at the 32nd International Conference on Machine Learning (ICML-MLOSS 2015), Lille, France July 10, 2015. ([http://d-scholarship.pitt.edu/26043/ paper][http://mloss.org/software/view/609/ code])
 * Huang, Y., González-Brenes, J. P., Kumar, R., Brusilovsky, P. (2015) A Framework for Multifaceted Evaluation of Student Models. In: Proceedings of the 8th International Conference on Educational Data Mining (EDM 2015), Madrid, Spain, pp. 203-210. ([http://www.educationaldatamining.org/EDM2015/uploads/papers/paper_164.pdf paper]) ([http://www.slideshare.net/huangyun/2015edm-a-framework-for-multifaceted-evaluation-of-student-models-polygon presentation])
 == Content Model Reduction for Better Learner Models ==

@@ Line 57: / Line 57: @@
 == Graph Analysis of Learner Models ==
 === Introduction ===
-This work explores the feasibility of a graph-based approach to model student knowledge in the domain of programming. The key idea of this approach is that programming concepts are truly learned not in isolation, but rather in combination with other concepts. Following this idea, we represent a student model as a graph where links are gradually added when the student's ability to work with connected pairs of concepts in the same context is con�rmed. We also hypothesize that with this graph-based approach a number of traditional graph metrics could be used to better measure student knowledge than using more traditional scalar models of student knowledge. To collect some early evidence in favor of this idea, we used data from several classroom studies to correlate graph metrics with various performance and motivation metrics.
+This work explores the feasibility of a graph-based approach to model student knowledge in the domain of programming. The key idea of this approach is that programming concepts are truly learned not in isolation, but rather in combination with other concepts. Following this idea, we represent a student model as a graph where links are gradually added when the student's ability to work with connected pairs of concepts in the same context is confirmed. We also hypothesize that with this graph-based approach a number of traditional graph metrics could be used to better measure student knowledge than using more traditional scalar models of student knowledge. To collect some early evidence in favor of this idea, we used data from several classroom studies to correlate graph metrics with various performance and motivation metrics.
 === Publications ===
 * Guerra, J., Huang, Y., Hosseini, R., & Brusilovsky, P. (2015, June). Graph Analysis of Student Model Networks. In Proceedings of the Second International Workshop on Graph-Based Educational Data Mining (GEDM 2015). CEUR-WS. ([https://d-scholarship.pitt.edu/secure/25933/1/graph_analysis.pdf paper]) ([http://www.slideshare.net/mallium/graph-analysis-of-student-model-networks presentation])

← Older revision		Revision as of 04:22, 5 April 2016
Line 60:		Line 60:

	=== Publications ===		=== Publications ===
		+	* Guerra, J., Huang, Y., Hosseini, R., & Brusilovsky, P. (2015, June). Graph Analysis of Student Model Networks. In Proceedings of the Second International Workshop on Graph-Based Educational Data Mining (GEDM 2015). CEUR-WS. ([https://d-scholarship.pitt.edu/secure/25933/1/graph_analysis.pdf paper]) ([http://www.slideshare.net/mallium/graph-analysis-of-student-model-networks presentation])

← Older revision		Revision as of 04:21, 5 April 2016
Line 54:		Line 54:
	* Gonzalez-Brenes, J. P., Huang, Y. The Leopard Framework: Towards understanding educational technology interventions with a Pareto Efficiency Perspective. In: The ICML 2015 Workshop on Machine Learning for Education (ICML 2015), Lille, France, 2015. ([https://dsp.rice.edu/sites/dsp.rice.edu/files/leopard_evaluation(1).pdf paper])		* Gonzalez-Brenes, J. P., Huang, Y. The Leopard Framework: Towards understanding educational technology interventions with a Pareto Efficiency Perspective. In: The ICML 2015 Workshop on Machine Learning for Education (ICML 2015), Lille, France, 2015. ([https://dsp.rice.edu/sites/dsp.rice.edu/files/leopard_evaluation(1).pdf paper])
	* Gonzalez-Brenes, J. P., Huang, Y. Using Data from Real and Simulated Learners to Evaluate Adaptive Tutoring Systems. In: 2nd AIED Workshop on Simulated Learners at the 17th Intl. Conf. on Artificial Intelligence in Education (AIED 2015), Madrid, Spain, 2015. ([http://ceur-ws.org/Vol-1432/sl_pap4.pdf paper])		* Gonzalez-Brenes, J. P., Huang, Y. Using Data from Real and Simulated Learners to Evaluate Adaptive Tutoring Systems. In: 2nd AIED Workshop on Simulated Learners at the 17th Intl. Conf. on Artificial Intelligence in Education (AIED 2015), Madrid, Spain, 2015. ([http://ceur-ws.org/Vol-1432/sl_pap4.pdf paper])
		+
		+	== Graph Analysis of Learner Models ==
		+	=== Introduction ===
		+	This work explores the feasibility of a graph-based approach to model student knowledge in the domain of programming. The key idea of this approach is that programming concepts are truly learned not in isolation, but rather in combination with other concepts. Following this idea, we represent a student model as a graph where links are gradually added when the student's ability to work with connected pairs of concepts in the same context is con�rmed. We also hypothesize that with this graph-based approach a number of traditional graph metrics could be used to better measure student knowledge than using more traditional scalar models of student knowledge. To collect some early evidence in favor of this idea, we used data from several classroom studies to correlate graph metrics with various performance and motivation metrics.
		+
		+	=== Publications ===

@@ Line 13: / Line 13: @@
 == Asymptotic Assessment of User Knowledge ==
-Asymptotic knowledge assessment is CUMULATE's legacy user modeling algorithm for computing user knowledge with respect to problem-solving. This is the current main learner model actively deployed in our educational systems. A more advanced one is an attempt to create a parameterizable version of this algorithm. [[CUMULATE asymptotic knowledge assessment|==> more]]
+Asymptotic knowledge assessment is CUMULATE's legacy user modeling algorithm for computing user knowledge with respect to problem-solving. This is the current main learner model actively deployed in our educational systems. A more advanced one has been proposed to create a parameterizable version of this algorithm. [[CUMULATE asymptotic knowledge assessment|==> more]]
 == Feature-Aware Student Knowledge Tracing (FAST) ==