Conference Paper

Building a knowledge repository of educational resources using dynamic harvesting

DOI: 10.1109/T4E.2010.5550041 Conference: Technology for Education (T4E), 2010 International Conference on
Source: IEEE Xplore


World Wide Web is hosting huge information regarding lots of areas and education is not an exception. Given the huge amount of data, searching for any educational resource manually is very difficult. To overcome this, an intelligent repository of educational resources that helps to decide among the available resources is needed. This paper discusses an attempt to build such repository. This will help users to decide among the available solutions for their needs by providing a comparative analysis among the solutions. The user will also be provided with user experience of the solutions. As the content over the web changes regularly and also new resources get added to the web, the repository will be updated dynamically. And all these tasks are done automatically as far as possible. This work uses crawling, classification, and information extraction techniques for the task of identifying the softwares/tools for education from the web. Our implementation focuses on the free open source softwares (FOSS) for education domain. The final framework of this system would be generic so that it can be extended to any other domain.

5 Reads
  • [Show abstract] [Hide abstract]
    ABSTRACT: This paper provides an overview of the advances on web mining in the domain of education by categorizing research on the field using the web-mining taxonomy: web content, web usage and web structure mining. Previous reviews focused on data mining methods applied to data derived from educational software applications (web based or not) and institutional administrative systems as well as from data gathered in typical classroom environments. The overview focuses on knowledge acquired from web based educational environments and the open web. More specifically it addresses: a) applications that dynamically update their content by extracting relevant educational information from the open web in order to meet user specific needs, b) applications that extract knowledge from usage data coming from educational web-based environments, and, c) applications that obtain knowledge coming from structure data such as links or social network connections that exist in e-learning applications and the open web.
    Proceedings of the 17th Panhellenic Conference on Informatics, Thessaloniki, Greece; 01/2013