Implementation of data mining techniques in elearning is a trending research area, due to the increasing
popularity of e-learning systems. E-learning systems provide
increased portability, convenience and better learning
experience. In this research, we proposed two novel schemes for
upgrading the e-learning portals based on the learner’s data for
improving the quality of e-learning. The first scheme is Learner
History-based E-learning Portal Up-gradation (LHEPU). In this
scheme, the web log history data of the learner is acquired. Using
this data, various useful attributes are extracted. Using these
attributes, the data mining techniques like pattern analysis,
machine learning, frequency distribution, correlation analysis,
sequential mining and machine learning techniques are applied.
The results of these data mining techniques are used for the
improvement of e-learning portal like topic recommendations,
learner grade prediction, etc. The second scheme is Learner
Assessment-based E-Learning Portal Up-gradation (LAEPU).
This scheme is implemented in two phases, namely, the
development phase and the deployment phase. In the
development phase, the learner is made to attend a short pretraining program. Followed by the program, the learner must
attend an assessment test. Based on the learner’s performance in
this test, the learners are clustered into different groups using
clustering algorithm such as K-Means clustering or DBSCAN
algorithms. The portal is designed to support each group of
learners. In the deployment phase, a new learner is mapped to a
particular group based on his/her performance in the pretraining program.