Exploring Machine Learning Methods to Automatically Identify Students in Need of Assistance

Ahadi, Alireza; Lister, Raymond; Haapala, Hannu; Vihavainen, Arto

doi:10.1145/2787622.2787717

Cited by 201 publications

(121 citation statements)

References 42 publications

Supporting

Mentioning

113

Contrasting

Unclassified

Order By: Relevance

“…Ahadi, Lister, Haapala, and Vihavainen (2015) explored machine learning techniques to analyse naturally accumulating programming process data (NAPD) to identify students in need of assistance. Similar data is analysed using principal component analysis in (Becker and Mooney, 2016) and here in section 4.…”

Section: Compiler Error Messagesmentioning

confidence: 99%

Effective compiler error message enhancement for novice programming students

Becker

Glanville

Iwashima

et al. 2016

Computer Science Education

View full text Add to dashboard Cite

Programming is an essential skill that all computing students must master. However programming can be difficult to learn. Compiler error messages are crucial for correcting errors, but are often difficult to understand and pose a barrier to progress for many novices. High frequencies of errors, particularly repeated errors, have been shown to be indicators of students who are struggling with learning to program. This study involves a custom IDE that enhances Java compiler error messages, intended to be more useful to novices than those supplied by the compiler. The effectiveness of this approach was tested in an empirical control/intervention study of approximately 200 students generating almost 50,000 errors. The design allows for direct comparisons between enhanced and non-enhanced error messages. Results show that the intervention group experienced reductions in the number of overall errors, errors per student, and several repeated error metrics. This work is important for two reasons. First, the effects of error message enhancement have been recently debated in the literature. This study provides substantial evidence that it can be effective. Second, these results should be generalizable at least in part, to other programming languages, students and institutions, as we show that the control group of this study is comparable to several others using Java and other languages.

show abstract

Section: Compiler Error Messagesmentioning

confidence: 99%

Effective compiler error message enhancement for novice programming students

Becker

Glanville

Iwashima

et al. 2016

Computer Science Education

View full text Add to dashboard Cite

show abstract

“…Other research concentrates on measuring the efficacy of tutoring software determining how robust learning is in an online tutor (Baker, Gowda, & Corbett, 2010;, knowledge that can then be fed back into the instructional design process. Ahadi, Lister, Haapala, and Vihavainen, (2015) outline a promising classifier (based on decision trees) approach to predicting low-performing and high-performing programming students. Based on a number of features, the most effective being how the students performed on a subset of Java programming exercises they were given during the course.…”

Section: Introductionmentioning

confidence: 99%

Utilizing student activity patterns to predict performance

Casey

Azcona

2017

Int J Educ Technol High Educ

View full text Add to dashboard Cite

Apart from being able to support the bulk of student activity in suitable disciplines such as computer programming, Web-based educational systems have the potential to yield valuable insights into student behavior. Through the use of educational analytics, we can dispense with preconceptions of how students consume and reuse course material. In this paper, we examine the speed at which students employ concepts which they are being taught during a semester. To show the wider utility of this data, we present a basic classification system for early detection of poor performers and show how it can be improved by including data on when students use a concept for the first time. Using our improved classifier, we can achieve an accuracy of 85% in predicting poor performers prior to the completion of the course.

show abstract

“…Purdue University (Course Signals) [20], University of Phoenix [21], and Capella University [22] are just a few examples of the universities that have utilized such systems. Logistic Regression has been a common method for the predictions [21,22], yet, more advanced methods from Machine Leaning have also been tried as well [23]. While Predictive Methods have been gaining more attention recently, the difficulty of accurately incorporating qualitative factors, such as student motivation and persistence which also influence student success, is still a major limitation for them.…”

Section: Related Workmentioning

confidence: 99%

Identifying Students’ Progress and Mobility Patterns in Higher Education Through Open-Source Visualization

Oran¹,

Martin²,

Klymkowsky³

et al. 2019

Preprint

View full text Add to dashboard Cite

The critical revision of the content, scope and alignment of curricula is essential for improving students success at Higher-Ed institutions. The effort is never trivial, as students are accepted from multiple sources (e.g. high schools, community colleges, other institutions) with different academic preparation, socioeconomic background, and motivation. In addition, students follow diverse paths either according to their interests, or according to the necessities, such as academic or financial requirements. Some graduate from their entry majors in 4 years, some need more time, some transfer to different disciplines, and some leave their universities. Accordingly, providing Higher-Ed decision makers with an accurate summary of these diverse student characteristics is a necessity to help them make better data-informed decisions for improving students success. In this regard, any data mining methodology that can convey valuable patterns from student data sets in clear and informative fashion will be valuable. In this study, we discuss the development and use of such a visual tool based on the Sankey Diagram. It presents students progress and mobility patterns in an easily understandable format, was developed using open source software, and was used by several departments of a research intensive Higher-Ed institution of more than thirty-thousand students during their academic review process. This paper provides a general discussion about how these visuals could be used in Higher-Ed institutions by discussing problems that can be addressed, detailing the dataneeds, the development methods, comparisons with other reporting methods, and how they were used in actual practice.

show abstract

Exploring Machine Learning Methods to Automatically Identify Students in Need of Assistance

Cited by 201 publications

References 42 publications

Effective compiler error message enhancement for novice programming students

Effective compiler error message enhancement for novice programming students

Utilizing student activity patterns to predict performance

Identifying Students’ Progress and Mobility Patterns in Higher Education Through Open-Source Visualization

Contact Info

Product

Resources

About