Automatic Extraction of Prerequisites Among Learning Objects Using Wikipedia-Based Content Analysis

Medio, Carlo De; Gasparetti, Fabio; Limongelli, Carla; Sciarrone, Filippo; Temperini, Marco

doi:10.1007/978-3-319-39583-8_44

Cited by 10 publications

(6 citation statements)

References 11 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Of particular relevance here is the line of related work, where researchers have investigated how machine learning and large-scale, open-access resources such as Wikipedia can be utilized to generate various types of educational content and interactions with the aim of scaling up computer-based learning systems and addressing the needs of their students (Brunskill et al, 2018;Dinan et al, 2018;Guo et al, 2016;Willis et al, 2019). In particular, it has been shown that the use of NLP techniques in application to Wikipedia may be helpful in generating pedagogically motivated concept maps to be used within an ITS (Lahti, 2009); identifying pre-requisite relations and sequencing among learning concepts to better model the learning path of the student and assess gaps in student's understanding of the subject (De Medio et al, 2016;Ramírez-Noriega et al, 2018;Talukdar & Cohen, 2012); and generating a variety of pedagogical interventions ranging from open questions Shah et al, 2017) to multiple-choice quizzes (Guo et al, 2016;Tamura et al, 2015) across a number of subject domains.…”

Section: Intelligent Tutoring Systemsmentioning

confidence: 99%

Automated Data-Driven Generation of Personalized Pedagogical Interventions in Intelligent Tutoring Systems

Kochmar

Belfer³

et al. 2021

Int J Artif Intell Educ

View full text Add to dashboard Cite

Intelligent tutoring systems (ITS) have been shown to be highly effective at promoting learning as compared to other computer-based instructional approaches. However, many ITS rely heavily on expert design and hand-crafted rules. This makes them difficult to build and transfer across domains and limits their potential efficacy. In this paper, we investigate how feedback in a large-scale ITS can be automatically generated in a data-driven way, and more specifically how personalization of feedback can lead to improvements in student performance outcomes. First, in this paper we propose a machine learning approach to generate personalized feedback in an automated way, which takes individual needs of students into account, while alleviating the need of expert intervention and design of hand-crafted rules. We leverage state-of-the-art machine learning and natural language processing techniques to provide students with personalized feedback using hints and Wikipedia-based explanations. Second, we demonstrate that personalized feedback leads to improved success rates at solving exercises in practice: our personalized feedback model is used in , a large-scale dialogue-based ITS with around 20,000 students launched in 2019. We present the results of experiments with students and show that the automated, data-driven, personalized feedback leads to a significant overall improvement of 22.95% in student performance outcomes and substantial improvements in the subjective evaluation of the feedback.

show abstract

Section: Intelligent Tutoring Systemsmentioning

confidence: 99%

Automated Data-Driven Generation of Personalized Pedagogical Interventions in Intelligent Tutoring Systems

Kochmar

Belfer³

et al. 2021

Int J Artif Intell Educ

View full text Add to dashboard Cite

show abstract

“…The identification of prerequisite relationships among Learning Objects has been addressed in [49], [50]. Learning Objects are associated to Wikipedia pages (topics), and their dependency is obtained using the classification of those topics supported by Wikipedia Miner [51].…”

Section: B Elicitation Of Relationships For Educational Purposesmentioning

confidence: 99%

A Combined Approach for Eliciting Relationships for Educational Ontologies Using General-Purpose Knowledge Bases

et al. 2019

View full text Add to dashboard Cite

Educational Ontologies describe learning domains by specifying the topics to be learned together with the relationships among the topics. Any technology supported learning system could take advantage of an educational ontology in the process of guiding students during their learning processes. This paper presents LiReWi, a system for the elicitation of relationships for educational ontologies from electronic textbooks. LiReWi combines grammar-based, co-occurrence-based and taxonomy-based methods together with several knowledge bases, such as Wikipedia, WordNet, WikiTaxonomy, WibiTaxonomy, and WikiRelations to elicit isA, partOf, prerequisite and pedagogicallyClose relationships between learning domain topics. LiReWi performs a three-step procedure to fulfill its task: first, all the topics are mapped to the diverse knowledge bases that will be used to identify the relationships; then, several relationship extractors, each using a different approach, are concurrently run to elicit candidate relationships; and, finally, the results are combined and filtered to obtain the final set of pedagogical relationships. LiReWi is designed following a modular approach that enables the inclusion of new relationship extractors. The evaluation conducted to validate the proposal is also reported in this paper. INDEX TERMS Technology supported learning systems, domain module, educational ontologies, relationship extraction, knowledge bases.

show abstract

“…Secondly, we annotated each set of transcripts, pertaining the same concept, by means of the Wikipedia Miner Toolkit. As a result we obtained, for each set of transcripts, a set of Wikipedia web pages, pertaining the same concept [3,9,11,8,10]. Then, we trained three binary learners: a decision tree, a naive-bayes and a multi-layer perceptron to inference whether a didactic relationship between the two LUs, both given in input, does exist or not.…”

Section: Motivations Goals and Related Workmentioning

confidence: 99%

Discovering Prerequisite Relationships Among Learning Objects: A Coursera-Driven Approach

Medio

Gasparetti

Limongelli

et al. 2016

Lecture Notes in Computer Science

Self Cite

View full text Add to dashboard Cite

In this work we address the problem of automatically finding prerequisite relations among learning materials in order to help instructional designers to speed up the course building process. Ours is a data-driven approach, where a (machine) learner is trained to classify predecessor/successor relationships, given two didactic materials in a textual form. As the training set we use the learning materials extracted from Coursera. A first evaluation shows promising results

show abstract

Automatic Extraction of Prerequisites Among Learning Objects Using Wikipedia-Based Content Analysis

Cited by 10 publications

References 11 publications

Automated Data-Driven Generation of Personalized Pedagogical Interventions in Intelligent Tutoring Systems

Automated Data-Driven Generation of Personalized Pedagogical Interventions in Intelligent Tutoring Systems

A Combined Approach for Eliciting Relationships for Educational Ontologies Using General-Purpose Knowledge Bases

Discovering Prerequisite Relationships Among Learning Objects: A Coursera-Driven Approach

Contact Info

Product

Resources

About