Research Of Personalized Learning Path Recommendation System Based On Online Learning Community

Online learning community learning is a form of learning with typical social characteristics.It is a development form of online learning to adapt to the era of fragmented learning.Because it has a convenient interactive communication mechanism and contains a large amount of learning resources,it provides a personalized education environment.In addition,the online learning community also stores a large number of learner behavior records left by learners.With the accumulation of data and resources,the so-called education "big data" has been gradually formed,which facilitates online learning.There are some problems that are difficult to solve:First,the emergence of education "big data" will lead to a significant increase in the likelihood of learners becoming trapped in learning and information overload;second,the separation of teachers and students in the online learning community,the lack of intervention by learners and The guidelines have made it difficult for learners to find a learning path that meets their actual needs,which has both undermined their enthusiasm for learning and reduced their learning efficiency.Third,online learning communities are unable to apply these data efficiently,and lack the means of learning support to adapt to students'characteristics.They are unable to translate the advantages of having big data for education into learning support services that satisfy learners,in order to support personalization in online learning communities.The implementation of education.In response to the above issues,this paper studies the learner model,domain knowledge model,and personalized learning path recommendation algorithm in the online learning community.The related work is as follows:Firstly,under the guidance of constructivist learning theory,the general process of path recommendation in online learning community is analyzed,and a research framework for personalized learning path recommendation applied to online learning community is designed.This framework can be used to guide online learning.Research and Practice of Personalized Learning Path Recommendation Service in Community.Second,the study of personalized learning paths recommends personalized learner models and domain knowledge models in online learning communities.By constructing a learner model,learning information left by learners can be fully utilized and quantified,and building a domain knowledge model can be used to describe knowledge in a structured way.The two work together to support a personalized learning path in an online learning community.The recommendation service went smoothly.Thirdly,the learning path recommendation algorithm based on the traditional ant colony algorithm is taken as the research object,and the important parameters,the pheromone renewal strategy,the local search strategy are mainly studied,and the improvement is proposed from three aspects:The first is the pheromone calculation accuracy optimization.Combining the characteristics of individualized learning paths,the learning path specific scoring guidelines and methods are studied,and a learning path scoring method based on the multi-factor fuzzy evaluation(FAHP)method is proposed.This method gives clear criteria for the learning path and effectively uses subjective and objective data.The qualitative learning review data is converted into quantitative learning path scoring data.The scoring data is used as the pheromone of the algorithm.It can solve the problem that the learner's subjective score is difficult to accurately represent the concentration of pheromone;secondly,the pheromone updating strategy is optimized and information is introduced.The existence of restriction intervals and restriction intervals provides the possibility of better exploration paths while ensuring the enlightenment of similar experience for ant colonies.Third,search strategy optimization is based on similar learners' "attracting" to each other and different learners' choices.The principle of mutual exclusion,using multiple ant colony parallelism Search,the introduction of heterogeneous learner repulsion factors affect the decision-making results,in order to effectively play the role of positive and negative feedback pheromone,improve the accuracy and degree of personalization of the recommended path.Fourth,take the "Data Structure" course in the online learning community as an example to conduct simulation experiments to achieve learning based on individual characteristics such as learner's learning goals,cognitive level,and resource preferences.The actual needs of the learning path,stimulate learning enthusiasm to improve learning results;and from a variety of indicators using multiple indicators of the experimental results to prove the feasibility and effectiveness of the proposed program.