A Green Cloud Service Provisioning Method for Mobile Micro-Learning

Li Yang, Ruijuan Zheng, Junlong Zhu, Mingchuan Zhang, Qingtao Wu

As a fundamental reform of traditional education, mobile micro-learning has developed rapidly in recent years. However, with the frequent access of users to cloud platforms, mobile terminal is facing serious energy consumption pressure, which limits the development of micro-mobile learning. Therefore, we propose a green cloud service provisioning method for mobile micro-learning. Firstly, the category homogenizing method and dynamic TF-IDF (D-TF-IDF) are used to classify user request, which can provide guidance for future service mode selection. Secondly, we search service from the 2-tier cloud architecture module according to the classified accuracy. Finally, Grey Wolf optimization (GWO) algorithm is used to find the server with the lowest energy consumption and finish service provisioning process. The simulation results demonstrate that our method can achieve saving energy goal. In addition, the accuracy of D-TF-IDF algorithm ran up to 83.91%, which is 2.92% higher than that of Nave Bayes algorithm, 3.65% than that of Rocchio algorithm, and 7.64% than that of TF-IDF algorithm.