LPR: an adaptive learning path recommendation system using ACO and meaningful learning theory
| dc.contributor.author | Niknam, Mehdi | |
| dc.contributor.examiningcommittee | Irani, Pourang (Computer Science) Matheos, Kathleen (Education) Madureira, Ana Maria (Computer Science, Polytechnic of Porto) | en_US |
| dc.contributor.supervisor | Thulasiraman, Parimala (Computer Science) | en_US |
| dc.date.accessioned | 2017-09-06T18:33:29Z | |
| dc.date.available | 2017-09-06T18:33:29Z | |
| dc.date.issued | 2017 | |
| dc.degree.discipline | Computer Science | en_US |
| dc.degree.level | Doctor of Philosophy (Ph.D.) | en_US |
| dc.description.abstract | In recent years, the educational community has been interested in enabled learning systems. That is, having a personalized learning system that can adapt itself while providing learning support to different learners to overcome the weakness of ‘one size fits all’ approaches in technology-enabled learning systems. In this thesis, we address one known problem in adaptive learning systems called curriculum sequencing. We design and implement a learning path recommendation (LPR) system that selects an appropriate learning path for learners based on their characteristics and needs. There are two components to the LPR system: searching for the learning paths and clustering the learners into groups based on their prior knowledge. Using bioinspired ant colony optimization (ACO) algorithm and meaningful learning theory of Ausubel, the ACO path finder component searches for a suitable learning path for the learner. This component incorporates continuous learner’s improvement in the process of a learning path selection. The LPR system, uses the pre-assessment/familiar degree calculator to gauge learner’s prior knowledge and produces a learner’s familiar degree of concepts. The clustering component uses Fuzzy C-Mean (FCM) algorithm. The LPR system can recommend more than one learning path to learners located on the cluster boundaries. We implement an interface to provide the recommendation to the learner. We evaluate the effectiveness of the LPR system by designing and developing a database course and ask actual learners to complete the course. The results of our experiment show that the group that used the LPR system have higher performance and knowledge improvement in the course than the control group. The performance and knowledge improvement differences between the two groups are statistically significant. Based on the statistical tests, the LPR system has a moderate to large impact on the learners’ performance and knowledge improvement. Although the course completion time for the LPR group was slightly less than the control group, no statistically significant difference is found between the time completion of both groups. | en_US |
| dc.description.note | October 2017 | en_US |
| dc.identifier.uri | http://hdl.handle.net/1993/32453 | |
| dc.language.iso | eng | en_US |
| dc.rights | open access | en_US |
| dc.subject | Ant Colony Optimization | en_US |
| dc.subject | Learning path recommendation | en_US |
| dc.subject | Content planning | en_US |
| dc.subject | Content sequencing | en_US |
| dc.subject | Fuzzy C-Mean Clustering | en_US |
| dc.title | LPR: an adaptive learning path recommendation system using ACO and meaningful learning theory | en_US |
| dc.type | doctoral thesis | en_US |