With the advancement of technologies, the usage of alternative eLearning systems as complementary
systems to the traditional education systems is becoming part of the everyday activities. At the same time, the creation of learning resources has increased exponentially
over time. However, the usability and reusability of these learning resources in various eLearning systems is difficult when they are unstandardised and semi-standardised learning
resources. Furthermore, eLearning activities’ lack of suitable personalisation of the overall learning process fails to optimize resources’ and systems’ potentialities. At the same time, the evolution of learning technologies and cloud computing creates new opportunities for
traditional eLearning to evolve and place the learner in the center of educational experiences.
This thesis contributes to a holistic approach to the field by using a combination of artificial intelligence techniques to automatically generate a personalized learning path for
individual learners using Cloud resources. We proposed an advancement of eLearning, named the Cloud eLearning, which recognizes that resources stored in Cloud eLearning can
potentially be used for learning purposes. Further, the personalised content shown to Cloud Learners will be offered through automated personalized learning paths. The main issue was to select the most appropriate learning resources from the Cloud and include them in a personalised learning path. This become even more challenging when these potential learning resources were derived from various sources that might be structured, semi- structure or even unstructured, tending to increase the complexity of overall Cloud eLearning retrieval and matching processes.
Therefore, this thesis presents an original concept,the Cloud eLearning, its Cloud eLearning Learning Objects as the smallest standardized learning objects, which permits reusing them because of semantic tagging with metadata. Further, it presents the Cloud eLearning Recommender System, that uses hierarchical clustering to select the most appropriate resources and utilise a vector space model to rank these resources in order of relevance for any individual learner. And it concludes with Cloud eLearning automated planner, which generates a personalised learning path using the output of the CeL recommender system.