Abstract:

A learning process consists of learning and verification activities. Each online lesson may be specified as a learning process. Usually, this learning process is linear, providing sequential execution of learning activities. All students, in this case, must use all learning activities and use the same learning material. This is a drawback of linear learning processes, as students are different, have different knowledge background and levels, have different learning objectives, motivations and learning styles.

To improve the efficiency of a learning process, it is necessary to take into account these differences among students. Learning is represented as a process consisting of learning activities and other events that may evaluate learning results or may allow students to communicate with their instructor or other students. A personalized e-learning process provides learning activities that are the most suitable for a student.

The personalization method should allow automatic generation of different learning paths of an online lesson, one for each category of students. Each learning activity provides an amount of knowledge suitable for one of the categories of students. Each evaluation activity provides the assessment of achieved results of learning. These results may be used to specify next, appropriate learning activity for each student. Instead of using evaluation results, a course designer may allow students to choose one of offered learning and other activities.

This allows a student to use not only a learning path of his category but also to mix learning paths of different categories. A process branch may be automatically activated as the result of knowledge assessment of a student or may be chosen by a student.

Use of fine-grained learning objects is one of the preconditions for implementation of the proposed personalization method. An atomic learning object may have one or more sections. Each section has three components: Section Title, Abstract, and Content, which holds the actual learning content in the form of text boxes, notes, figures, code snippets. A complex learning object contains many atomic learning objects. An online lesson usually includes both complex and atomic learning objects.

BMU has developed mDita Editor as an authoring tool for development of multimedia knowledge presentations of e-learning materials. BMU is using mDita Editor for creation of its e-learning materials for almost 200 online courses, and its Repository of learning objects currently contains more than 70.000 developed learning objects. Metadata are used for searching of learning objects, when an author needs to find a suitable learning object, already developed, for its online lesson. Use of the Repository of developed learning objects increases their reusability. Reusability of learning objects enhances the efficiency of the creation of online lessons, reduces their cost, and improves their quality.

The Repository of Learning Objects contains DITA objects. mDita Server is the software module of the developed systems that performs these conversions and manages developed DITA objects in their Repository.

The proposed approach to personalization of e-learning in this paper allows that online students in a class may use different online learning materials and different learning paths.

Keywords:

e-learning, personalized e-learning, adaptable e--learning processes, learning objects, learning processes, personalization of learning.