Presenting a personalized mobile learning recommender system by using environmental and location information

Abstract

The most important aims of mobile learning systems is providing relations between the real and virtual worlds and creating related instructing strategies with education activities in order to obtain appropriate educational achievements in addition to creating motivations in using mobile systems among the learning individuals. Offering potentials of instructing activities and places interested by individuals in this regard are among the most important aspects by which learners could increase their knowledge levels. In this respect and by describing the case, we offer a model in the form of a mobile learning recommender system, abbreviated by "PMLES". Hence in this article we try to improve the performance of recommender systems in mobile learning environments by presenting an integrated system and providing a mobile learning recommender system, and along with that offer a learning method appropriate with the level of knowledge among different learners. The aim for this proposing model is increasing the performance rate of recommender systems by considering the educational conditions and developments of each learner in intercity environments. The learning contents should be determined with appropriation to the person's needs and conditions, and on the basis of the goals for the learning system. Therefore, regarding the diverse applications of recommender systems and mobile learning, these systems could be used in environmental sciences and the matter could focus around the subjects including geology, mining, surveying, geography and other relevant topics due to non-urban environment in this proposed system. When, in a non-urban environment, a learner needs instruction information, the system should recommend proper instruction information to him/her that are appropriate with the attending place of that individual, by using allocating tools such as GPS and according to the moving direction and location of the person. It means that the learner should be within such a system and the instructions and suggestions to him/her should improve the learning trends of the person. The learning recommender system is presented in the first section. Then a definition of the features of PMLR system will be given. To follow, the modeling procedures of PMLRS and the states and architecture of this system will be considered. Afterwards, the PMLRS (personalized mobile learning recommender system) driver is described and finally, the obtained results are given. II. PERSONALIZED MOBILE LEARNING RECOMMENDER SYSTEM Constructivist theory and situated learning, as the two types of instruction theories should always be considered in learning recommender systems. The first theory expresses the notion that the learner always combines his/her knowledge and past experiences or understandings with his/her up-to-date activities and experiences to obtain new knowledge. Accordingly, the performance of the person is in trial and error basis and his/her feedback in each stage provides the possibility of obtaining the best output. The second theory is dealing with the surrounding world, sociability of individuals, social activities and obtained experiences or the acquired knowledge in that respect. It indicates that this theory is clearly related to the existing themes in the real world and their relevant problems. It is the physical attendance of people that enables them to enhance and reinforce their environmental information and learning about related subjects with the considered area. The proposed combined recommender system method is established from combination of the methods based on the learners' behavioral patterns, using the recommender algorithms in classifying the learners, and also identifying the dependence of instruction contents with the environmental characteristics of features of education space in promoting the level of knowledge in learners. By dividing the learners into different groups, this system deals with considering the performance and behaviors of them and then recommends subjects or proper tests for them and regarding the obtained level of knowledge of each individual tries to offer the most appropriate educating place. The advantage of this method is in classification of the learners with regards to their level of knowledge. Hence after entering of a new learner, the system will be able to predict their functions