The notion of 'orchestrating learning' has gained acceptance within the TEL research community in recent years. However, there is little consensus about what 'orchestration' means, and what orchestrating learning in a concrete educational context entails. This paper aims to address these two concerns. Through a literature review focused especially on the field of TEL, we provide definitions of this orchestration, and gather the most commonly cited aspects of orchestration into a unified conceptual framework. This emergent framework is then used as an analytical lens to structure data from an existing case study in order to illustrate its usefulness as a tool to understand and propose new solutions to aid orchestration in complex, real-world TEL situations. Although further theorisation and modelling of orchestration is still needed, the presented framework provides a first step, backed up by a serious review of the field.

Blended learning models that combine face-to-face and online learning are of great importance in modern higher education. However, their development should be in line with the recent changes in e-learning that emphasize a student-centered approach and use tools available on the Web to support the learning process. This paper presents research on implementing a contemporary blended learning model within the e-course “Hypermedia Supported Education”. The blended model developed combines a learning management system (LMS), a set of Web 2.0 tools and the E-Learning Activities Recommender System (ELARS) to enhance personalized online learning. As well as incorporating various technologies, the model combines a number of pedagogical approaches, focusing on collaborative and problem-based learning, to ensure the achievement of the course learning outcomes. The results of the comparative study show the effectiveness of the proposed model in that students who performed personalized collaborative e-learning activities achieved better course results. These findings encourage the further application of the model to other computer science courses.

This paper presents the results of a comparative research of paper-based testing versus online testing using MudRi. The research was conducted on three courses at the Department of Informatics, University of Rijeka, and included 1231 tests. In observed courses, online tests have largely replaced traditional paper-based tests in the last few years. Therefore, the aim of this research was to determine whether the online testing results differ from the results of traditional paper-based tests. The research results showed that, for the observed courses, the median values of the results achieved during MudRi online tests and traditional paper based tests are not significantly different when online tests include certain percentage of subjective test questions.

The paper presents the work-in-progress with the aim to develop recommender system for personalization of activities in e-learning 2.0 environment. The main components of the proposed system are activity, student and group models, and recommender module. Activity model will be used for learning design representation and will include items that could be recommended to students: e-tivities, possible collaborators, tools, and advices. To provide recommendations tailored to the student's and group's characteristics, an important component of the system will include student and group models. The emphasis of the research is on the procedures for assessing the student's (group's) activity level based on the data collected from the third party services (Web 2.0 tools). Student's model will also represent knowledge level and preferences. The recommender module, as third component of the system, will include original pedagogical rules together with the algorithms that adapt known recommendations techniques to the educational context.