Research On Supported Model And Implementing Mechanism For Adaptive Learning System

Some web-based learning systems, such as Blackboard, Moodle, SaKai, etc, provide thesame learning resources to learners. According to user characteristics, personalized dynamicadaptive learning content don’t presented. So the construction of adaptive learning system isan effective scheme to resolve the problem between dynamic students’ needs of resources andstatic teaching resources. It is also an effective method to resolve the contradiction betweenoverall infinite resources and individual needs of finite resources. It realizes two functions inthe distance learning environment, which one is that learers can actively adapt distancelearning, organize learning resources, make and carry out learning plan, select study strategyand can be self-assessment.Another is that system can provide learners with adaptive learningservice support and in this learning circumstance, the system will put the students as servicecustomers, as god and provide learning methods, strategies and emotional service.That is tosay, adaptive learning system can recommend personal learning path and learning resources,according to students’ characteristics and learning behaviors, such as learning styles, mediatypes, learning interests, cognitive level, etc. It can effectively enhance students’ individualeducation and improve the full development of a growing personality.Adaptive learning system is an ascendant research direction. It is possible that part ofachievements has been widely used. However, the construction of system model is still veryweak and this system also lacks the necessary empirical research. This thesis attempts topromote this research from model construction and teaching practice, and also applies variousresearch methods, for example literature research, interpretative structural modelling method,empirical research, investigation research, evaluative method, etc. The following is thecontents of this work.Firstly, this thesis analyses the necessity of studying adaptive learning system throughliterature research and investigation research. Then, we propose and construct AdaptiveLearning System which can provide personal learning resource and learning paths based-oncommon features and insufficiences of current adaptive learning systems.Secondly, based on the features and insufficiences of current adaptive learning systemsreference model, this thesis proposes a reference model of Adaptive Learning System,including user model, domain model (or knowledge model), adaptation model, adaptiveengine, presentation model, and adding learning behaviors description and adaptivepresentation description of learning resources, navigation and sequence in favor of developingsystem and defining system guidelines. Thirdly, based-on learning style and knowledge level, this thesis designs user modelconstruction reference standard by using interpretative structural modelling method. Thisthesis adopts two steps to construct learning style model, which first has been initializedthrough Felder-Silverman learning style model, and then has been updated based-on rules orBayesian Nets method. Knowledge level model has been constructed based-on Cumulativescore law concept and cognitive theory of Blum.Fourthly, this thesis designs knowledge model construction reference standard by usinginterpretative structural modelling method and ontology technology, which provides aprerequisite for adaptively presenting knowledges based-on user model.Fifthly, using interpretative structural modelling method, this thesis designs some objectmodels and relation models, such as chapter and section, section and knowledge, knowledgeand learning object, which can recommend personal learning path and learning resourcesaccording to students’ characteristics and learning behaviors, such as learning styles, mediatypes, learning interests, cognitive level, etc.Sixthly, this thesis designs adaptive learning system architecture and diversified learningresources based-on double-subject personalized recommendation strategy. Then, based-onobject model and relation model this thesis has developed Adaptive Learning System by usingJAVA, SQLServer and ontology technology, which is set as an example to introduce howlearning objects and learning active sequence have been represented based-on user model, andintroduce personal optimizational recommendation of learning active sequence based-onAprioriAll algorithm, which students can avoid learning lost, reduce cognitive overload andthe learning efficiency and quality can been improved.Seventhly, based-on Adaptive Learning System, this thesis analyzes user behaviorsobjective variables and evaluates learning effectiveness from user-centered evaluations.Theresults show that adaptive and personalized learning can improve learning effectiveness andstudents can complete the learning tasks using less time and learning objects. Meanwhile,students give a positive on learning enjoyment, learning effort, learning motivation andlearning satisfaction in adaptive learning system.According to research progress and continual reflection, this thesis proposes that thereare still some points which need to be studied in further. For example, we may studypersonalization recommendation among learners based-on collaborative filtering, which ishelpful that students can find new knowledge and adaptive learning system may be applied toother subject teaching and basic education learners to evaluate learning effectiveness and howadaptive learning achievement may be migrated from E-learning to U-learning, which can beused effectively.