Unlocking the Potentials of Mobile Learning Object Compilation by Using Random Forest and Semantic Web Based as Tool for Lecturers

Abstract

Mobile learning, is an area which make use of moveable devices to get access to learning and its activities. Regrettably, most universities and establishments in developing countries and beyond do not have learning contents that are compatible with the mobile devices. This makes the development of mobile learning contents difficult and therefore the practice of reusing online Learning Objects (LO), is generally employed to make the development of the mobile learning contents much easier. The major issues encountered with semantic web at times of processing RLO, are that first, not all online LO are accessible on mobile devices, subsequently the LO metadata are not readily available. The Mobile Learning Objects Compilation Framework (MLOC) which is a mixture framework of random forest and semantic web is proposed by this study to address these problems.   The hybrid framework would include a method that will generate RLO metadata from repositories and use those metadata to evaluate the RLO. The assembled related RLO are made available to these learning contents to other systems through the web services so that mobile apps can access the RLO easily. This research therefore examines the methods to enhance semantic web in the reuse of LO for mobile devices. The research will first introduce a method to generate learning metadata from public search results based on learning theories.  Thereafter, establish the semantic methods to evaluate the RLO and assemble RLO into complete learning units in a repository that can be accessed by mobile devices without hitches  The projected framework would be able to search and extract the RLOs which are much more effectual compared to RLOs retrieved by other related mobile apps which in turn confirms that MLOC can be used to process RLOs for mobile devices.