Improving the Quality of Mobile Learning Services

Abstract

The quality of services (QoS) has a key role to improve the performances of the collaborative mobile learning. However, many factors can limit it such as the cache management policies, the collaboration architecture, a heavy communication, etc. Therefore, we are interested in developing an impeccable learning system supporting cache management, collaboration architecture decisiveness, based mobile agent and web services standardization. The aim is to improve the quality of mobile learning services in term of information availability (hit rates and byte hit rates), learners’ average latency and the cognitive profiles of students. Consequently, we implemented four removal policies named Least Frequently Used (LFU), First In First Out (FIFO), Least Recently Used (LFU) and Size. In addition, we proposed and evaluated three collaboration architectures based mobile agent named Sequential, Random and diffusion mechanism. Experimental results show that our system improves the cognitive profiles of students and reduces the learners’ average latency. Furthermore, the experience and the performance results gained from this system prove that the removal policies performances depend on the learning method of each student and its collaboration patterns. For this raison, it is advantageous to allow the student to choose and/or change the removal policy during each learning session. However, diffusion mechanism provides the best performances in term of learners’ average latency.