Abstract | ||
---|---|---|
In recent years, learning analytics solutions have highly appealed to the higher education community who mainly focuses on improving the learning process, self-regulated learning skills, and learners' success rate. Learning analytics has to deal with continuous data, however, conventional data mining algorithms are not readily applicable to handle the continuous incoming of learners' data. In order to cope with these scenarios, the proposed learning analytics aimed to manage the continuous data, perform the clustering process using the optimization approach, detect the 'at-risk' learners' who are in a course failure situation, and generate signals to learners and teachers. Based on the predicted outcome, the proposed system identifies and adapts the learning activities and learning contents to help learners find their way out of their learning difficulties and course failure situation. The experiments were conducted to analyze the performance of the proposed work using the simulated learners' data. The experimental results provide empirical evidence that the proposed work reduces the course failure rate and improves learners' success rate. |
Year | DOI | Venue |
---|---|---|
2015 | 10.1109/T4E.2015.14 | T4E |
Keywords | Field | DocType |
learning analytics, big data, parallel particle swarm optimization clustering, Naive Bayes prediction, recommendation system, learner's competence, big data | Semi-supervised learning,Algorithm design,Empirical evidence,Learning analytics,Computer science,Failure rate,Artificial intelligence,Cluster analysis,Analytics,Multimedia,Machine learning,Higher education | Conference |
ISSN | Citations | PageRank |
2372-7217 | 0 | 0.34 |
References | Authors | |
17 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kannan Govindarajan | 1 | 150 | 13.37 |
Vivekanandan Suresh Kumar | 2 | 16 | 7.04 |
David Boulanger | 3 | 11 | 5.38 |
Kinshuk | 4 | 2123 | 389.31 |