Research On Multi-dimensional Multi-granular Learner Personality Model And Its Processing Strategies

With the rapid development of new information technology,such as wireless network technology,mobile terminal technology,cloud computing,artificial intelligence etc.it has promoted the emergence of a variety of digital learning theory and practice,from distance learning,to e-learning,to mobile learning,to ubiquitous adaptive learning,etc.The traditional online learning system provides the same learning plan,learning path and learning resources for all learners.The online learning system is difficult to follow the learner’s learning objectives,cognitive level,learning style,learning interests,etc,to set up the learning programs and recommended learning resources.As a result,it leads learners to studying in a poor efficiency,such as lack of personality,single scence,etc.On the other hand,it is easy to appear the problems like"learning lost","information overload",etc.when learners face massive learning resources.To solve these problems,adaptive U-learning system becomes a new research direction of online learning system.Knowledge model,learner personality model and teaching object interaction model,as the core module of Adaptive U-learning system,has become a research hotspot.According to the learner’s personality,adaptive U-learning system provides accurate learning services for learners includes learning resources,learning plan,learning path,etc,so that it can improve quality and efficiency,reduce the cost of learning,promote the reform of education informationization based on mobile Internet.Aiming at the common problems of the learner personality model,such as the information of personality description insufficiency,the perspective of single character description,low integration of model,etc.,this thesis proposes learning multi-dimensional multi-granular personality model-LPM~3 and its processing strategies.We research maintenance strategy and model specification of the model.In order to enhance the stability of interaction,the interactive processing strategy based on connected graph is proposed for ubiquitous learning interaction between objects.The following is the contents of this thesis.(1)This thesis has analyzed and discussed in detail the key personality attributes which influences and decides the learning efficiency of the learner on the foundation of refering to domestic and international learner personality model.Personality attributes include enrollment information,learning objectives,knowledge level,learning style,learning trace,learning interests.This thesis has analyzed the relationship between them in detail and then put forward the multi-dimensional multi-granularity learner personality model.It has discussed the operation mechanism of multi-dimension multi-granularity,the structure design,the multi-dimensional and multi granularity,and the data model specification of multi dimension multi granularity learner model.(2)Based on the multi-dimensions multi-granularity characteristics of LPM~3 model,this thesis proposes explicit acquisition strategies,implicit acquisition strategies and inference acquisition strategies to obtain different personality attributes.It puts forward to maintenance strategy based on point-in-time and time window,maintenance strategies based on linkage.(3)This thesis proposes a state interaction model for ubiquitous learning between teachers and students,learners and learners,which have a poor interaction except the internet.Based on directed connected graph to give an interaction state judgment strategies,by finding the connected paths to find the interaction path,by finding the shortest path to find the optimal path of the interaction,and puts forward based on connected graph interaction processing strategy IPSCG to optimal interactive performance find the simplest path dredge.Experimental results show that these strategies can decrease the interaction costs,improve the interactive performance,and enhance the stability of the user interaction.It makes a tangible contribution between teachers and students,learners and learners,learners and learning resources for their interaction and the sharing of resources in a stable and efficient way.The results of this study can provide personalized learning knowledge service for learners effectively,reduce knowledge trek,solve the cognitive overload and resources insufficient phenomenon,stimulate interest in learning,and improve learning efficiency.It works on all-round development of the personality of learners and improve the level of modernization of education,promote balanced development of education.The results of this study has a good practical value.With the development of wearable devices and big data technology,the next step we will research:Firstly,we will assimilate learner situations,emotional and other personal property into multi-dimensional multi-granularity learner personality model.We will study the influence of these properties and the relationship between attributes.Secondly,we study the construction of personality characteristics of the dynamic changes in the dimension of flexible multi-dimensional multi-granular learner personality model,and then provide learners with a better quality of personalized learning services.