Research On Learning Path Recommendation And Evaluation Method Based On Intelligent Collaborative Mechanism

In recent years,with the continuous progress of technology,online learning systems have become more complete.The education industry has gradually entered a more segmented field,mainly reflected in diverse ways of expression,more open resources,and more intelligent content provided.At the same time,online education platforms have flooded with a large number of users,but this is also accompanied by problems such as low learner efficiency,information overload,and knowledge loss.In view of this,an intelligent collaborative mechanism based learning path recommendation and evaluation method was studied by analyzing learners’ behavior and learning process data,and an intelligent learning system was designed and developed based on this.The specific content is as follows:Firstly,In order to express the relationship between knowledge points and provide learners with a more logical learning order,this is takes junior high school mathematics as the research object to build a map of junior high school mathematics domain knowledge.Acquire the knowledge information related to junior high school mathematics needed to build the map through data collection technology,use data pre-processing technology to rationalize all data sources,and then use knowledge extraction and knowledge fusion technology to identify entities and expand and update the map.For the constructed map,the quality of the map is evaluated through the identification of domain experts,so as to complete the construction of junior high school mathematics knowledge map,And store it in a graphical database as data support for subsequent research.Secondly,In response to the problem of knowledge loss that learners encounter when learning from massive resources,this paper proposes a multi algorithm collaborative personalized learning path recommendation method.This method constructs a learner model by analyzing the learner’s behavior data on the online learning platform.According to the junior high school mathematics domain knowledge map,it generates a sequence of knowledge points for each learner through the data mining algorithm,which is based on which swarm intelligence algorithm is used to provide adaptive learning strategies for learners,so as to achieve personalized learning path recommendation.This article has designed a large number of simulation experiments from different perspectives,and the experimental results show that using multi algorithm collaboration to recommend learning paths for learners on online learning platforms is feasible.Finally,The knowledge state and learning state of learners are a dynamic process during learning.In order to capture the learning situation of learners during the learning process,this paper designs a learning process evaluation model based on multidimensional features.Collect learning process log data of learners on online learning platforms,use feature extraction methods to obtain learners’ feature data,analyze learners’ features from multiple dimensions,concatenate all features,input them into the learning process evaluation model,and output the final results.Adjust learners’ learning plans in a timely manner based on changes in learners’states.After extensive simulation experiments,multiple dimensions of features were considered and inputted into the network model to evaluate the learning process.Compared to other models,the performance of the model is better,and the model can effectively evaluate the state changes of learners during the learning process.Based on the above research on learning path recommendation and learning process evaluation,this article designs and implements an intelligent learning system.The accuracy,effectiveness,and interpretability of the proposed intelligent algorithm based learning path recommendation and evaluation method are verified through modules such as path recommendation,process evaluation,and learning resources,opening up new ideas for the provision of personalized services in online learning platforms.