Kaleidoscope home
A centralized research data repository
helploginprintemailchat   log in  
  Search  

Welcome to the centralized research data repository.

Our Kaleidoscope Initiative project, called the "Centralised Data Repository", is focused on the problem of how researchers can share educational data and compare empirical results. Every learning science project tends to have its own, idiosyncratic data and data format, sometimes taking a high-level perspective, sometimes a low-level perspective, but this is highly problematic when researchers later decide to compare results, share data, and collaborate more closely with one another. For instance, how can researchers retrieve comparable objects from different databases? How can two or more researchers unify their work without totally rewriting one (or more) databases? Our objective in this project was to explore ways in which we could overcome these problems, proposing a way to make educational data common and accessible to a wide variety of educational researchers.

We propose to tackle the problem by developing a software mediator (see Figure 1), that accesses a common ontology (see Figure 2), representing a wide range of possible learning objects.

The mediator architecture
Figure 1: The Mediator architecture

The mediator's role is to map the common ontology (see Figure 2) to the individual ontologies specific to each of project and to provide common access to all data.

Beta version of the common ontology.
Figure 2: Beta version of the common ontology. Please click here to see the enlarged picture

In other words, we would like to provide one "portal" or query language that can be used to access the data in a range of projects. To design such a solution, we have teamed with a variety of researchers both within and outside of the Kaleidoscope network, including:

  • Jean-Michel Adam, Jean-Pierre David, ARCADE, Laboratoire CLIPS, Fédération Grenoble, Université Joseph Fourier;
  • Andreas Harrer, Dept. of Computer Science, Universität Duisburg;
  • Alejandra Martínez Monés, Departamento de Informática, Universidad de Valladolid;
  • Ton de Jong, Faculty of Behavioral Sciences, University of Twente;
  • Ryan Baker, Carnegie Mellon University.

The approach we took was to obtain data from the educational technology systems developed and used by the above researchers and create a common ontology that could be mapped to all of the individual ontologies. We also reused and extended an architecture developed on our own ActiveMath project as the basis for the mediator design proposed in the project.

NEW: All the ontologies, including the common one, are now downladable!

As part of our Kaleidoscope Symposium presentation we will:

  • Introduce and discuss the problem of variable educational data;
  • Describe the general mediator architecture we propose to address the problem;
  • Present the general ontology we have devised, including examples of how the general ontology maps to specific project ontologies;
  • Discuss how this approach could be realized and implemented in a full-scale project;
  • Discuss the benefits we eventually hope to achieve through this effort.
  • Contribute with your data format to the Centralized Repository!