Automatic Concept Extraction for Domain and Student Modeling in Adaptive Textbooks

Chau, Hung; Labutov, Igor; Thaker, Khushboo; He, Daqing; Brusilovsky, Peter

doi:10.1007/s40593-020-00207-1

Cited by 23 publications

(12 citation statements)

References 41 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…These approaches are not directly applicable to our problem setting since we require short phrases focusing on the performance of an action by the user. Approaches such as TextRank (Kazemi et al, 2020), SingleRank (Wan and Xiao, 2008), Ex-pandRank (Wan and Xiao, 2008), TopicRank (Bougouin et al, 2013), TopicalPageRank (Sterckx et al, 2015), PositionRank (Florescu and Caragea, 2017), Bi-LSTM-CRF Sequence Labeling (Alzaidy et al, 2019), FACE (Chau et al, 2020), and Mul-tipartiteRank (Boudin, 2018) have been applied to the task of key phrase extraction from documents as opposed to theme detection from clusters. For cluster-labeling, there has been work around using keyword extraction from clusters, utilizing WordNet synsets to expand the keywords, followed by a selection procedure to assign the final label (Poostchi and Piccardi, 2018;Chang and McKeown, 2019).…”

Section: Related Workmentioning

confidence: 99%

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track

2022

View full text Add to dashboard Cite

It has been a privilege chairing the Industry Track this year. We thank the conference general chair, Dan Roth, for inviting us to the organizing committee. Thanks also to Program Chairs Marine Carpuat, Marie-Catherine de Marneffe and Ivan Vladimir Meza Ruiz, and all members of the organizing committee. Yunayo Li and Owen Rambow served as advisors for the Industry Track to provide continuity for this new track. We were generously helped by every member of this committee over the past year, and organizing this track was possible only with their advice and efforts. We once again recognize and thank every member of the industry track program committee for volunteering their time. Finally, thanks to the authors and attendees of the industry track for embracing this initiative and offering a reason to continue the industry track at NAACL-HLT conferences.

show abstract

Section: Related Workmentioning

confidence: 99%

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track

2022

View full text Add to dashboard Cite

show abstract

“…Extracting KCs from textbooks has been explored before for the domain of information retrieval. For instance, Chau and colleagues (Chau et al 2020) adopted a supervised machine learning approach based on a set of expert-defined features. The features they used fall into three broad categories: linguistic, positional, or statistical.…”

Section: Related Workmentioning

confidence: 99%

“…All ngrams (n = 1...4) are considered candidate keyphrases. We used this n-gram range for extracting candidate key concepts based on a quick analysis of the textbook by ourselves as well as based on literature, e.g., Chau and colleagues (Chau et al 2020) define domain concepts as "single words or short phrases of two to four words." The candidate n-grams contain stop words and punctuation marks.…”

Section: Candidate Concept Generationmentioning

confidence: 99%

“…Domain modeling is the task of specifying the units of knowledge, also called Knowledge Components (KCs), in a target domain such as physics, biology, or computer programming. A domain model includes a structure that specifies the relationship among the KCs, typically in the form of a prerequisite knowledge structure suggesting a specific trajectory towards mastery, i.e., a particular order in which students should master the KCs (Goldin, Pavlik Jr, and Ritter 2016;Koedinger, Corbett, and Perfetti 2012;Chau et al 2020). Domain model can also link the KCs to specific learning activities or objects that allow learners to master those KCs through practice.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Domain Model Discovery from Textbooks for Computer Programming Intelligent Tutors

Banjade

Oli

Tamang³

et al. 2021

FLAIRS

View full text Add to dashboard Cite

We present a novel approach to intro-to-programming domain model discovery from textbooks using an over-generation and ranking strategy. We first extract candidate key phrases from each chapter in a Computer Science textbook focusing on intro-to-programming and then rank those concepts according to a number of metrics such as the standard tf-idf weight used in information retrieval and metrics produced by other text ranking algorithms. Specifically, we conduct our work in the context of developing an intelligent tutoring system for source code comprehension for which a specification of the key programming concepts is needed - the system monitors students' performance on those concepts and scaffolds their learning process until they show mastery of the concepts. Our experiments with programming concept instruction from Java textbooks indicate that the statistical methods such as KP Miner method are quite competitive compared to other more sophisticated methods. Automated discovery of domain models will lead to more scalable Intelligent Tutoring Systems (ITSs) across topics and domains, which is a major challenge that needs to be addressed if ITSs are to be widely used by millions of learners across many domains.

show abstract

“…There are designs of models of affective users that consider the emotional states of the student, that is, that allow the probabilistic combination of information about the causes and effects of emotional reactions (Conati & Maclaren, 2009). Also, the creation of a domain model is fundamental in an adaptive hypermedia system, since it de nes the relationships between concepts within an adaptive educational environment (Chau, Labutov, Thaker, He, & Brusilovsky, 2020). The domain model must guarantee the presentation of these contents in various formats, considering the different presentation styles of the domain model; all these elements will contribute to the success of the experience (Howlin & Lynch, 2014).…”

Section: Related Wordsmentioning

confidence: 99%

Adaptive hypermedia system in times of COVID-19: Application in Management Accounting course 1 at the University of Havana.

López

Vidal

Balmaseda

2021

Preprint

View full text Add to dashboard Cite

New trends in higher education, accentuated by the global COVID19 pandemic, visualize the need for more flexible virtual courses. The Faculty of Accounting and Finance of the University of Havana, analyzes new technological configurations that facilitate the personalization of the educational process. In this context, this work aims to design an adaptive hypermedia system prototype applied to the Management Accounting 1 course at the University of Havana to guarantee the delivery, distribution, and variety of learning content to students in a personalized way. An empirical research supported by an experimental method is presented, with a sample of 94 students, a Control Group made up of 41 students and an Experimental Group made up of 53 students; Descriptive statistics and the Pareto principle support the results. It is concluded that, 1. The personalization of the educational process through the application of adaptive techniques leads to academic success at an individual and group level, in this sense, the Experimental group only needed a final exam session compared to the three calls of the group of control; 2. Need to continue investigating new adaptation criteria, in this sense, the mean scores obtained by the Experimental group did not have a significant increase in relation to that of the Control group. This work also contributes to visualizing educational equity referring to the capacity of the educational system to offer special supports that guarantee that educational objectives are achieved in a broader dimension; innovation is related to technologies and new strategies to optimize online educational processes.

show abstract

Automatic Concept Extraction for Domain and Student Modeling in Adaptive Textbooks

Cited by 23 publications

References 41 publications

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track

Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies: Industry Track

Domain Model Discovery from Textbooks for Computer Programming Intelligent Tutors

Adaptive hypermedia system in times of COVID-19: Application in Management Accounting course 1 at the University of Havana.

Contact Info

Product

Resources

About