MURAL - Maynooth University Research Archive Library



    VEAP: a visualisation engine and analyzer for preSS#


    Culligan, Natalie and Quille, Keith and Bergin, Susan (2016) VEAP: a visualisation engine and analyzer for preSS#. In: Koli Calling '16 Proceedings of the 16th Koli Calling International Conference on Computing Education Research. ACM, pp. 130-134. ISBN 9781450347709

    [img]
    Preview
    Download (511kB) | Preview


    Share your research

    Twitter Facebook LinkedIn GooglePlus Email more...



    Add this article to your Mendeley library


    Abstract

    Computer science courses have been shown to have a low rate of student retention. There are many possible reasons for this, and our research group have had considerable success in pinpointing the factors that influence outcome when learning to program. The earlier we are able to make these predictions, the earlier a teacher can intervene and provide help to an at-risk student, before they fail and/or drop out. PreSS (Predict Student Success) is a semi-automated machine learning system developed between 2002 and 2006 that can predict the performance of students on an introductory programming module with 80% accuracy, after minimal programming exposure. Between 2013 and 2015, a fully automated web-based system was developed, known as PreSS#, that replicates the original system but provides: a streamlined user interface; an easy acquisition process; automatic modeling; and reporting. Currently, the reporting component of PreSS# outputs a value that indicates if the student is a "weak" or "strong" programmer, along with a measure of confidence in the prediction. This paper will discuss the development of VEAP: a Visualisation Engine and Analyser for PreSS#. This software provides a comprehensive data visualisation and user interface, that will allow teachers to view data gathered and processed about institutions, classes and individual students, and provides access to further user-defined analysis, to allow a teacher to view how an intervention could influence a student's predicted outcome.

    Item Type: Book Section
    Keywords: Computer science; Education; Data Visualization ; Educational Tools;
    Academic Unit: Faculty of Science and Engineering > Computer Science
    Item ID: 10307
    Identification Number: https://doi.org/10.1145/2999541.2999553
    Depositing User: Dr. Susan Bergin
    Date Deposited: 11 Dec 2018 15:21
    Publisher: ACM
    Refereed: Yes
    URI:

    Repository Staff Only(login required)

    View Item Item control page

    Downloads

    Downloads per month over past year