LONET: An interactive search network for intelligent lecture path generation

  • Authors:
  • Neil Y. Yen;Timothy K. Shih;Qun Jin

  • Affiliations:
  • Waseda University, Fukushima, Japan;National Central University, Taiwan (R.O.C.);Waseda University, Saitama, Japan

  • Venue:
  • ACM Transactions on Intelligent Systems and Technology (TIST) - Special section on agent communication, trust in multiagent systems, intelligent tutoring and coaching systems
  • Year:
  • 2013

Quantified Score

Hi-index 0.00

Visualization

Abstract

Sharing resources and information on the Internet has become an important activity for education. In distance learning, instructors can benefit from resources, also known as Learning Objects (LOs), to create plenteous materials for specific learning purposes. Our repository (called the MINE Registry) has been developed for storing and sharing learning objects, around 22,000 in total, in the past few years. To enhance reusability, one significant concept named Reusability Tree was implemented to trace the process of changes. Also, weighting and ranking metrics have been proposed to enhance the searchability in the repository. Following the successful implementation, this study goes further to investigate the relationships between LOs from a perspective of social networks. The LONET (Learning Object Network), as an extension of Reusability Tree, is newly proposed and constructed to clarify the vague reuse scenario in the past, and to summarize collaborative intelligence through past interactive usage experiences. We define a social structure in our repository based on past usage experiences from instructors, by proposing a set of metrics to evaluate the interdependency such as prerequisites and references. The structure identifies usage experiences and can be graphed in terms of implicit and explicit relations among learning objects. As a practical contribution, an adaptive algorithm is proposed to mine the social structure in our repository. The algorithm generates adaptive routes, based on past usage experiences, by computing possible interactive input, such as search criteria and feedback from instructors, and assists them in generating specific lectures.