Front Page
RESEARCH TOPICS
Automatic Essay Assessment
Concretizing Technologies in Special Education
Development Project for Technology Education
Information and Communication Technology for Development
Kids' Club
Mobile Learning
Seniors' Club
Software Visualization & Jeliot
Technologies for Children with Individual Needs
Text and Program Analysis
Virtual Studies of Computer Science
WHO & WHAT
News @ EdTech
People @ EdTech
Mémoire d'un Étudiant Français
Publications @ EdTech
IMPDET
University of Joensuu Department of Computer Science and Statistics @ University of Joensuu Academy of Finland European Commission European Social Fund TEKES - National Technology Agency of Finland SIGCSE - ACM Special Interest Group in Computer Science Education Assa Abloy Connexor Discendum Heureka Karjalainen Lingsoft Microsoft Nokia Sordino Information Systems
CONFERENCES, WORKSHOPS & SUMMER SCHOOLS
IEEE 4th International Workshop on Technology for Education in Developing Countries, July 10 -12, 2006, Iringa, Tanzania Malta International Summer School on Educational Technology in Cultural Context, June 20 - 25, 2005 Taiwan International Summer School on Educational Technology in Cultural Context, Taiwan, June 28 - July 3, 2005 International Conference on Educational Technology in Cultural Context, Taiwan, July 7, 2005 Mekrijärvi International Summer School on Educational Technology in Cultural Context, Finland, September 25 - 29, 2005 See more...
TEXT AND PROGRAM ANALYSIS (TAPAS)
TAPAS is a research consortium between Education Technology reasearch group (EdTechΔ) at the Department of Computer Science at the University of Joensuu (UJ) and the Computer Science Education research group (COMPSER) the Laboratory of Software Technology at the Helsinki University of Technology (HUT).
In the project, we aim at combining the
essay assessment expertise of
EdTechΔ group with the knowledge of COMPSER group about the assessment environments. The goal of the project is to develope methods and tools for automated free text and program text analysis.
In this project three directions are pursued.
Firstly, we plan to develop methods for free text analysis that are context independent.
Secondly, we combine these free text analysis methods with analysis methods of program texts
in order to
evaluate programming assignments. Thirdly, the developed tools are evaluated in realistic situations in programming courses. The first part of the project is mainly carried out byt the EdTechΔ group in UJ and in the second phase we aim at combining the approaches from the two institutions to assess programming assignments or projects. The empirical evaluation is then carried out at HUT and other partner institutions.
The text analysis methods is designed from the following aspects that are interrelated. For example, an assessment method is always based on a chosen measure to analyze a text.
- Novel assessment methods will be developed for the content and structural analysis of texts. The critical factors include time efficiency as well as accuracy, i.e., the correlation between the grades given by a human assessor and the grading method.
- Semi-automated assessment features are needed in order to give formative feedback to teachers and students. These features will open an essay in a "white-box", relating the evaluation and grading to the criteria of assessment. Semi-automatic features such as automated feedback and summary generation can therefore support the learner during the writing process, not only after it.
- Versatile measures and representations thereof will be developed for the quality of a text. Besides the conventional numeric grade for the contents of an essay, the writing style could be evaluated in terms of structure, coherence, and cohesion. Argumentation, conceptual weaknesses, or sudden changes in vocabulary could be represented visually.
To implement a smart environment for automated assessment, three dimensions of an essay needs to be analyzed: the learning dimension to indicate the cognitive level of the learning process, the linguistic dimension characterizing the structure of the essay, and the technical dimension to address available or needed techniques for assessing the essay throughout the writing process. Without a reasonable model for essay assessment, one can never end up with an intelligent assessment environment that would adapt to the many needs of essay evaluation, in each dimension. One reason behind the limitations of essay assessment techniques is the lack of a sufficient model which would cover the three aspects, which co-exist in any essay grading situation, whether manual or automatic. Thus, the three dimensions form a basis for an architecture of a formative semi-automatic essay evaluation and assessment environment.
This research project is funded by Academy of Finland (2006-2009).
Tuomo Kakkonen, Niko Myller, Jari Timonen and Erkki Sutinen, firstname.lastname@cs.joensuu.fi