Open Corpus Personalized Learning - Revision history

K.thaker at 16:34, 11 January 2023

2023-01-11T16:34:45Z

K.thaker at 15:59, 11 January 2023

2023-01-11T15:59:18Z

K.thaker at 15:46, 11 January 2023

2023-01-11T15:46:36Z

K.thaker at 15:46, 11 January 2023

2023-01-11T15:46:13Z

K.thaker at 15:31, 11 January 2023

2023-01-11T15:31:17Z

K.thaker at 15:25, 11 January 2023

2023-01-11T15:25:31Z

Peterb: /* Publications */

2023-01-10T17:18:18Z

‎Publications

Peterb: /* Publications */

2023-01-10T17:15:48Z

‎Publications

Peterb: /* Publications */

2023-01-10T17:15:35Z

‎Publications

H.chau: /* The Project Team */

2017-05-12T23:02:06Z

‎The Project Team

@@ Line 1: / Line 1: @@
 Goal: This project challenges the assumption that adaptive hypermedia systems require expensive knowledge engineering for domain and content modeling. It replaces carefully-crafted domain model with automatically-created domain models, lowering the cost of developing adaptive educational hypermedia software while also providing a wider range of instructional paths through the content. Adaptive educational hypermedia is known for its ability to improve learning outcomes and engagement maximizing educational opportunity for learners with different levels of knowledge. The development of this more automatic, open-corpus approach to adaptive educational hypermedia will increase the volume and the variety of resources available for meaningful online learning, especially for individuals learning on their own. Automatic knowledge indexing of educational content makes the system easy to maintain and update over time. These new open corpus user modeling techniques automatically adapt user models and personalized guidance to new materials as they are acquired. The ability to automatically organize, index, and adaptively recommend distributed educational content without the need of manual processing by system developers, enables new material to be integrated dynamically and with minimal effort in response to student needs.
-This project merges research on text analysis, human learning, and personalization to enable open corpus personalized learning. It develops its models of the domain and human learning from an initial set of well-organized, manually selected materials. Automatic text analysis creates an ensemble of domain models with different characteristics. Each individual model may be flawed or incomplete, however collectively they provide comprehensive coverage of the topic from several perspectives, thus reducing the manual effort required to create adaptive educational hypermedia. Multiple perspectives also give the system more flexibility in how to guide each student. These domain models are used as a foundation for building and maintaining dynamic models of user knowledge. The ensemble of domain and user models is used to deliver reactive and proactive adaptive guidance in an open corpus context. The growth of a person's knowledge is inferred by observing learner behavior and obtaining occasional feedback. This exploratory research opens the way to open corpus personalized learning. The domain modeling, user modeling, and personalization techniques developed in this research will be evaluated using a multi-layer framework that includes assessment by subject experts, performance prediction, cross-validation, and user studies.
+This project merges research on text analysis, human learning, and personalization to enable open corpus personalized learning. It develops its models of the domain and human learning from an initial set of well-organized, manually selected materials. Automatic text analysis creates an ensemble of domain models with different characteristics. Each individual model may be flawed or incomplete, however, collectively they provide comprehensive coverage of the topic from several perspectives, thus reducing the manual effort required to create adaptive educational hypermedia. Multiple perspectives also give the system more flexibility in how to guide each student. These domain models are used as a foundation for building and maintaining dynamic models of user knowledge. The ensemble of domain and user models is used to deliver reactive and proactive adaptive guidance in an open corpus context. The growth of a person's knowledge is inferred by observing learner behavior and obtaining occasional feedback. This exploratory research opens the way to open corpus personalized learning. The domain modeling, user modeling, and personalization techniques developed in this research will be evaluated using a multi-layer framework that includes assessment by subject experts, performance prediction, cross-validation, and user studies.
 ==The Project Team==
@@ Line 23: / Line 23: @@
 Digitization has brought a significant change to the educational sector in the past ten years, making digital textbooks a convenient and popular resource for learning. Starting as merely a digital copy of paper textbooks, digital textbooks gradually integrated a number of innovative and advanced functionality. Rapid digitization also remarkably increased the volume of open learning resources, which have grown very popular, providing learners with great opportunities to expand their learning. Moreover, these new digital learning resources could be combined with the traditional learning process. For example, a learner may come across the concept of ``Indexing'' while reading a textbook on ``Information Retrieval''. This triggers her to engage extra online materials on ``Indexing'' to learn the state-of-the-art practical methods or research trends.
-The goal of the project is to address this challenge by providing links to related content at different textbooks material. In this project we would like to use concepts as a backbone for connecting different education materials. This is a challenging task since textbooks and online materials are written by  different authors, for different audiences, and frequently use different terms/words to express the same concepts. Furthermore, many domains lack formal domain models or ontologies, where all domain concepts are listed and organized. Therefore, the association of terms and concepts, and the representation of concepts should be automatically discovered and constructed.
+The goal of the project is to address this challenge by providing links to related content at different textbook materials. In this project, we would like to use concepts as a backbone for connecting different education materials. This is a challenging task since textbooks and online materials are written by different authors, for different audiences, and frequently use different terms/words to express the same concepts. Furthermore, many domains lack formal domain models or ontologies, where all domain concepts are listed and organized. Therefore, the association of terms and concepts, and the representation of concepts should be automatically discovered and constructed.
-To address these challenges, our work applies distributed neural representations, which demonstrated an ability to solve the term-mismatch problem by transferring the representation of terms to a N-dimension continuous vector (also called embedding). The expectation is that terms or words which share the same semantic meaning will be nearer in this N-dimensional continuous vector space. However, these approaches cannot be applied efficiently to educational resources, as they need a good deal of training data to learn efficient representations (for example vectors learned from entire Google News Corpus), while the volume of educational resources for less popular domains might be limited. In addition, these embeddings do not bring insights to topical distribution of terms, for example ``Precision'', ``Recall'', ``Cross-validation'' are concepts related to same topic ``Evaluation'' in ``Machine Learning'' domain.
+To address these challenges, our work applies distributed neural representations, which demonstrated an ability to solve the term-mismatch problem by transferring the representation of terms to an N-dimension continuous vector (also called embedding). The expectation is that terms or words which share the same semantic meaning will be nearer in this N-dimensional continuous vector space.
 |}

@@ Line 13: / Line 13: @@
 Image:Jordan.jpeg|[http://pitt.edu/~jab464 Jordan Barria-Pineda]
 Image:Avatar.jpg|[http://pitt.edu/~hkc6 Hung Chau]
-Image:k.thaker.png|[http://pitt.edu/~kmt81 Khushboo Thaker]
+Image:k.thaker.png|[http://kthaker.com Khushboo Thaker]
 </gallery>

@@ Line 19: / Line 19: @@
 ==Education Content Linking==
 {|
-|valign="top" | [[ Image:Content-linking.png|thumb|left|'''100'''|Backbone connecting Educational Systems]]
+|valign="top" | [[ Image:Content-linking.png|thumb|left|'''100'''|Concepts Backbone connecting Educational Systems]]
 |valign="center" |
 Digitization has brought a significant change to the educational sector in the past ten years, making digital textbooks a convenient and popular resource for learning. Starting as merely a digital copy of paper textbooks, digital textbooks gradually integrated a number of innovative and advanced functionality. Rapid digitization also remarkably increased the volume of open learning resources, which have grown very popular, providing learners with great opportunities to expand their learning. Moreover, these new digital learning resources could be combined with the traditional learning process. For example, a learner may come across the concept of ``Indexing'' while reading a textbook on ``Information Retrieval''. This triggers her to engage extra online materials on ``Indexing'' to learn the state-of-the-art practical methods or research trends.

← Older revision		Revision as of 15:46, 11 January 2023
Line 16:		Line 16:

	</gallery>		</gallery>
		+
		+	==Education Content Linking==
		+	{\|
		+	\|valign="top" \| [[ Image:Content-linking.png\|thumb\|left\|'''100'''\|Backbone connecting Educational Systems]]
		+	\|valign="center" \|
		+	Digitization has brought a significant change to the educational sector in the past ten years, making digital textbooks a convenient and popular resource for learning. Starting as merely a digital copy of paper textbooks, digital textbooks gradually integrated a number of innovative and advanced functionality. Rapid digitization also remarkably increased the volume of open learning resources, which have grown very popular, providing learners with great opportunities to expand their learning. Moreover, these new digital learning resources could be combined with the traditional learning process. For example, a learner may come across the concept of ``Indexing'' while reading a textbook on ``Information Retrieval''. This triggers her to engage extra online materials on ``Indexing'' to learn the state-of-the-art practical methods or research trends.
		+
		+	The goal of the project is to address this challenge by providing links to related content at different textbooks material. In this project we would like to use concepts as a backbone for connecting different education materials. This is a challenging task since textbooks and online materials are written by different authors, for different audiences, and frequently use different terms/words to express the same concepts. Furthermore, many domains lack formal domain models or ontologies, where all domain concepts are listed and organized. Therefore, the association of terms and concepts, and the representation of concepts should be automatically discovered and constructed.
		+
		+	To address these challenges, our work applies distributed neural representations, which demonstrated an ability to solve the term-mismatch problem by transferring the representation of terms to a N-dimension continuous vector (also called embedding). The expectation is that terms or words which share the same semantic meaning will be nearer in this N-dimensional continuous vector space. However, these approaches cannot be applied efficiently to educational resources, as they need a good deal of training data to learn efficient representations (for example vectors learned from entire Google News Corpus), while the volume of educational resources for less popular domains might be limited. In addition, these embeddings do not bring insights to topical distribution of terms, for example ``Precision'', ``Recall'', ``Cross-validation'' are concepts related to same topic ``Evaluation'' in ``Machine Learning'' domain.
		+	\|}
		+

	==Knowledge Extraction==		==Knowledge Extraction==

@@ Line 19: / Line 19: @@
 ==Knowledge Extraction==
 {|
-|valign="top" | [[ Image:Knowledge-linking-Illustration.png|thumb|left|'''100'''|Dynamic Knowledge Modeling in Textbook Reading]]
+|valign="top" | [[ Image:Knowledge-linking-Illustration.png|thumb|left|'''100'''|Dynamic Linking Textbooks]]
 |valign="center" |
 The Internet has dramatically increased both the volume and variety of online educational resources, such as online textbooks, online courses, and tutorials. The development of modern search techniques has further promoted the quick access of these resources. However, most of these educational resources are not well-structured, which imposed an important challenge -- readers without sufficient background knowledge may be difficult to understand its content. To achieve the goal of recommending ''the right content'' that matches individuals' knowledge levels, the first critical step is to provide a better organization for educational resources. The project visions two important components when organizing educational resources: (1) knowledge concept extraction; and (2) concept hierarchy extraction. Traditional solutions for these two problems heavily rely on experts' manual efforts which are time-consuming and unscalable.

@@ Line 63: / Line 63: @@
 * Meng, R., Zhao, S., Han, S., He, D., Brusilovsky, P., and Chi, Y. (2017) Deep Keyphrase Generation. In:  Proceedings of ACL2017, Annual Meeting of the Association for Computational Linguistics, Vancouver, Canada, July 30-August 4, 2017 pp. 582-592.
 * Labutov, I., Huang, Y., Brusilovsky, P., and He, D. (2017) Semi-Supervised Techniques for Mining Learning Outcomes and Prerequisites. In:  Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada, ACM, pp. 907-915.
-* Thaker, K., Huang, Y., Brusilovsky, P., and He, D. (2018) Dynamic Knowledge Modeling with Heterogeneous Activities for Adaptive Textbooks�. In:  Proceedings of the 11th International Conference on Educational Data Mining, Buffalo, USA, July 15-18, 2018, pp. 592-595.
+* Thaker, K., Huang, Y., Brusilovsky, P., and He, D. (2018) Dynamic Knowledge Modeling with Heterogeneous Activities for Adaptive Textbooks. In:  Proceedings of the 11th International Conference on Educational Data Mining, Buffalo, USA, July 15-18, 2018, pp. 592-595. [http://educationaldatamining.org/files/conferences/EDM2018/papers/EDM2018_paper_199.pdf paper]
 * Thaker, K. M., Brusilovsky, P., and He, D. (2018) Concept Enhanced Content Representation for Linking Educational Resources. In:  Proceedings of 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI), Santiago, Chile, December 3-6, 2018, IEEE, pp. 413-420.
 * Barria-Pineda, J., Brusilovsky, P., and He, D. (2019) Reading Mirror: Social Navigation and Social Comparison for Electronic Textbooks. In:  Proceedings of First Workshop on Intelligent Textbooks at 20th International Conference on Artificial Intelligence in Education (AIED 2019), Chicago, USA, June 25, 2019, CEUR, pp. 30-37.

@@ Line 67: / Line 67: @@
 * Barria-Pineda, J., Brusilovsky, P., and He, D. (2019) Reading Mirror: Social Navigation and Social Comparison for Electronic Textbooks. In:  Proceedings of First Workshop on Intelligent Textbooks at 20th International Conference on Artificial Intelligence in Education (AIED 2019), Chicago, USA, June 25, 2019, CEUR, pp. 30-37.
 * Thaker, K., Brusilovsky, P., and He, D. (2019) Student Modeling with Automatic Knowledge Component Extraction for Adaptive Textbooks. In:  Proceedings of First Workshop on Intelligent Textbooks at 20th International Conference on Artificial Intelligence in Education (AIED 2019), Chicago, USA, June 25, 2019, CEUR, pp. 95-102.
-Thaker, K., Zhang, L., He, D., and Brusilovsky, P. (2020) Recommending Remedial Readings Using Student’s Knowledge State. In:  Proceedings of 13th International Conference on Educational Data Mining, July 10-13, 2020, pp. 233-244.
+* Thaker, K., Zhang, L., He, D., and Brusilovsky, P. (2020) Recommending Remedial Readings Using Student’s Knowledge State. In:  Proceedings of 13th International Conference on Educational Data Mining, July 10-13, 2020, pp. 233-244.
 * Yuan, X., Wang, T., Meng, R., Thaker, K., Brusilovsky, P., He, D., and Trischler, A. (2020) One Size Does Not Fit All: Generating and Evaluating Variable Number of Keyphrases. In: D. Jurafsky, J. Chai, N. Schluter and J. R. Tetreault (eds.) Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online, July 5-10, 2020, Association for Computational Linguistics, pp. 7961-7975.
 * Chau, H., Labutov, I., Thaker, K., He, D., and Brusilovsky, P. (2021) Automatic Concept Extraction for Domain and Student Modeling in Adaptive Textbooks. International Journal of Artificial Intelligence in Education  31 (4), 820–846.
 * Wang, M., Chau, H., Thaker, K., Brusilovsky, P., and He, D. (2022) Knowledge Annotation for Intelligent Textbooks. Technology, Knowledge and Learning, in press.

← Older revision		Revision as of 17:15, 10 January 2023
Line 61:		Line 61:
	* Huang, Yun and Yudelson, Michael and Han, Shuguang and He, Daqing and Brusilovsky, Peter. "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://d-scholarship.pitt.edu/28248/ paper]).		* Huang, Yun and Yudelson, Michael and Han, Shuguang and He, Daqing and Brusilovsky, Peter. "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://d-scholarship.pitt.edu/28248/ paper]).
	* Meng, Rui and Han, Shuguang and Huang, Yun and He, Daqing and Brusilovsky, Peter. "Knowledge-based Content Linking for Online Textbooks." In Proceeding of the 2016 IEEE/WIC/ACM International Conference on Web Intelligence, pp. 13-16. IEEE Computer Society, 2016. ([http://d-scholarship.pitt.edu/30486/1/wi16-knowledge-linking.pdf paper]).		* Meng, Rui and Han, Shuguang and Huang, Yun and He, Daqing and Brusilovsky, Peter. "Knowledge-based Content Linking for Online Textbooks." In Proceeding of the 2016 IEEE/WIC/ACM International Conference on Web Intelligence, pp. 13-16. IEEE Computer Society, 2016. ([http://d-scholarship.pitt.edu/30486/1/wi16-knowledge-linking.pdf paper]).
		+	* Meng, R., Zhao, S., Han, S., He, D., Brusilovsky, P., and Chi, Y. (2017) Deep Keyphrase Generation. In: Proceedings of ACL2017, Annual Meeting of the Association for Computational Linguistics, Vancouver, Canada, July 30-August 4, 2017 pp. 582-592.
		+	* Labutov, I., Huang, Y., Brusilovsky, P., and He, D. (2017) Semi-Supervised Techniques for Mining Learning Outcomes and Prerequisites. In: Proceedings of the 23rd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, Halifax, NS, Canada, ACM, pp. 907-915.
		+	* Thaker, K., Huang, Y., Brusilovsky, P., and He, D. (2018) Dynamic Knowledge Modeling with Heterogeneous Activities for Adaptive Textbooks�. In: Proceedings of the 11th International Conference on Educational Data Mining, Buffalo, USA, July 15-18, 2018, pp. 592-595.
		+	* Thaker, K. M., Brusilovsky, P., and He, D. (2018) Concept Enhanced Content Representation for Linking Educational Resources. In: Proceedings of 2018 IEEE/WIC/ACM International Conference on Web Intelligence (WI), Santiago, Chile, December 3-6, 2018, IEEE, pp. 413-420.
		+	* Barria-Pineda, J., Brusilovsky, P., and He, D. (2019) Reading Mirror: Social Navigation and Social Comparison for Electronic Textbooks. In: Proceedings of First Workshop on Intelligent Textbooks at 20th International Conference on Artificial Intelligence in Education (AIED 2019), Chicago, USA, June 25, 2019, CEUR, pp. 30-37.
		+	* Thaker, K., Brusilovsky, P., and He, D. (2019) Student Modeling with Automatic Knowledge Component Extraction for Adaptive Textbooks. In: Proceedings of First Workshop on Intelligent Textbooks at 20th International Conference on Artificial Intelligence in Education (AIED 2019), Chicago, USA, June 25, 2019, CEUR, pp. 95-102.
		+	Thaker, K., Zhang, L., He, D., and Brusilovsky, P. (2020) Recommending Remedial Readings Using Student’s Knowledge State. In: Proceedings of 13th International Conference on Educational Data Mining, July 10-13, 2020, pp. 233-244.
		+	* Yuan, X., Wang, T., Meng, R., Thaker, K., Brusilovsky, P., He, D., and Trischler, A. (2020) One Size Does Not Fit All: Generating and Evaluating Variable Number of Keyphrases. In: D. Jurafsky, J. Chai, N. Schluter and J. R. Tetreault (eds.) Proceedings of the 58th Annual Meeting of the Association for Computational Linguistics, Online, July 5-10, 2020, Association for Computational Linguistics, pp. 7961-7975.
		+	* Chau, H., Labutov, I., Thaker, K., He, D., and Brusilovsky, P. (2021) Automatic Concept Extraction for Domain and Student Modeling in Adaptive Textbooks. International Journal of Artificial Intelligence in Education 31 (4), 820–846.
		+	* Wang, M., Chau, H., Thaker, K., Brusilovsky, P., and He, D. (2022) Knowledge Annotation for Intelligent Textbooks. Technology, Knowledge and Learning, in press.