G.P. Papamichail (Greece) and D.P. Papamichail (USA)
Concept creation, k-means range clustering, Personalizeddecision making, Supervised e-learning.
This paper describes a dynamic clustering approach for real-time concept creation in distributed e-learning systems. Firstly the individual learner through a preference definition stage effectively narrows the feature space according to her learning objectives. Then an efficient version of the k-means algorithm, namely the k means range, is employed for the creation of learning concept clusters. These two stages, the pre-processing orthogonal range search followed by the dynamic clustering, enhance the intuitive transformation of large data sets into meaningful clusters. The complexity analysis of the aforementioned algorithm demonstrates the real-time computational efficiency of this research approach. Finally the potential and implications on distributed e-learning and teaching are discussed.
Important Links:
Go Back
