Explaining Individual and Collective Programming Students’ Behavior by Interpreting a Black-Box Predictive Model

Pereira, Filipe Dwan; Fonseca, Samuel C.; Oliveira, Elaine; Cristea, Alexandra I.; Bellhäuser, Henrik; Rodrigues, Luiz; Oliveira, David Braga Fernandes de; Isotani, Seiji; Carvalho, Leandro Silva Galvão de

doi:10.1109/access.2021.3105956

“…Um dos resultados mais atuais sobre a predic ¸ão de desempenho é a explicabilidade do modelo de aprendizagem de máquina [Pereira et al 2021a]. O grupo de pesquisa entende que uma resposta numérica obtida a partir de métodos caixa preta não atende aos anseios do professor e do aluno.…”

Section: Algumas Iniciativas E Resultadosunclassified

Learning Analytics no Ensino de Introdução à Programação de Computadores

Oliveira

¹

,

Oliveira²,

Carvalho³

et al. 2022

Anais Do I Workshop De Aplicações Práticas De Learning Analytics Em Instituições De Ensino No Brasil (WAPLA 2022)

0

View full text Add to dashboard Cite

Disciplinas de programação introdutória apresentam uma alta taxa de reprovação no mundo todo, e na Universidade Federal do Amazonas, isso também acontece. Desde que um grupo de professores dessa instituição resolveu reformular a disciplina, algumas iniciativas de Learning Analytics vem sendo adotadas. O objetivo deste artigo e apresentá-las de maneira abrangente, além de alguns resultados obtidos durante esses últimos anos de pesquisa.

show abstract

“…Moreover, this human support would help the AI to improve its rules to and to become even more precise in annotating topics. Furthermore, such human/AI interaction is claimed by the literature [13,24,38] to be dominant in modern systems. The reason is that humans and AI have different strengths and weaknesses, and the combination of the heterogeneous intelligence of both agents can be quite powerful.…”

Section: Applications and Implicationsmentioning

confidence: 99%

Towards Supporting Cs1 Instructors and Learners With Fine-Grained Topic Detection in Online Judges

Pereira¹,

Fonseca²,

Wiktor³

et al. 2022

Preprint

1

0

View full text Add to dashboard Cite

<p>Online judges (OJs) are a popular tool to support programming learning. However, one major issue with OJs is that problems are often put together without any associated meta-information that could, for example, be used to help classify problems. This meta-information could be extremely valuable to help users quickly find what types of problems they need most. To face this problem, several OJ administrators have recently begun manually annotating the topics of problems based on computer science-related subjects, such as dynamic programming, graphs, and data structures. Initially, the topics were used to support programming competitions and experienced learners. However, with OJs increasingly used to support CS1 classes, such topic annotation needs to be extended to suit CS1 learners and instructors. In this work, for the first time, to the best of our knowledge, we propose and validate a predictive model that can automatically detect fine-grained topics of problems based on the CS1 syllabus. After experimenting with many shallow and deep learning models with different word representations based on cutting-edge NLP techniques, our best model is a CNN, achieving an F1-score of 88.9%. We then present how our model can be used for various applications, including (i) facilitating the search process of problems for CS1 learners and instructors and (ii) how it can be integrated into a system to recommend problems in OJs.</p>

show abstract

“…The use of online judges in education to evaluate students' programs has brought many advantages [54,44,56,5,26,20,38,43]. First, the instructor could assess the correctness of students' submissions more efficiently and accurately.…”

Section: Online Judgesmentioning

confidence: 99%

Towards Supporting Cs1 Instructors and Learners With Fine-Grained Topic Detection in Online Judges

Pereira¹,

Fonseca²,

Wiktor³

et al. 2022

Preprint

1

0

View full text Add to dashboard Cite

<p>Online judges (OJs) are a popular tool to support programming learning. However, one major issue with OJs is that problems are often put together without any associated meta-information that could, for example, be used to help classify problems. This meta-information could be extremely valuable to help users quickly find what types of problems they need most. To face this problem, several OJ administrators have recently begun manually annotating the topics of problems based on computer science-related subjects, such as dynamic programming, graphs, and data structures. Initially, the topics were used to support programming competitions and experienced learners. However, with OJs increasingly used to support CS1 classes, such topic annotation needs to be extended to suit CS1 learners and instructors. In this work, for the first time, to the best of our knowledge, we propose and validate a predictive model that can automatically detect fine-grained topics of problems based on the CS1 syllabus. After experimenting with many shallow and deep learning models with different word representations based on cutting-edge NLP techniques, our best model is a CNN, achieving an F1-score of 88.9%. We then present how our model can be used for various applications, including (i) facilitating the search process of problems for CS1 learners and instructors and (ii) how it can be integrated into a system to recommend problems in OJs.</p>

show abstract

Explaining Individual and Collective Programming Students’ Behavior by Interpreting a Black-Box Predictive Model

Abstract: This work involved human subjects or animals in its research. The authors confirm that all human/animal subject research procedures and protocols are exempt from review board approval.

Cited by 36 publications

References 66 publications

Learning Analytics no Ensino de Introdução à Programação de Computadores

Learning Analytics no Ensino de Introdução à Programação de Computadores

Towards Supporting Cs1 Instructors and Learners With Fine-Grained Topic Detection in Online Judges

Towards Supporting Cs1 Instructors and Learners With Fine-Grained Topic Detection in Online Judges

Contact Info

Product

Resources

About