Research On Learning Path Recommendation Strategy Based On Knowledge Map

With the rise of artificial intelligence and big data technology,domain knowledge map building technology and resource recommendation technology have gradually become a research hotspot,which are widely used in the business field.In recent years,information technology in the field of education continues to penetrate,online learning has become a common phenomenon,knowledge system construction and learning path recommendation technology has also been accelerated development,so there are more choices for learners.However,in the vast amount of network resources,people often to solve a small problem and access to a large number of unrelated information costs precious time,which is prone to information overload or learning lost.At present,the academic research on the learning path recommendation system has made some achievements,but there are still many deficiencies if it is to be applied to practice.It is mainly manifested in: the accuracy of the learning path recommended by the system to learners is not very high,which leads to low learning efficiency of learners;the lack of reasonable and effective and scientific domain knowledge model construction;the fuzzy weakening of online learners' personality in order to improve the efficiency of the system;the research on the learning path recommended by the way of learners' questions is still rare and immature.In view of the above shortcomings,this paper studies the learning path recommendation strategy based on knowledge map from the perspective of junior high school mathematics.The following is the main work of this paper:(1)The knowledge model and the complex relationship between knowledge points are studied,and the knowledge map of junior high school mathematics field is constructed.(2)Based on the actual needs of learners-asking questions,according to the questions asked by learners and the weakness of the results of job analysis in the operating system,a feature weighted naive Bayesian classifier is designed for problem classification,through which the target knowledge point objects to be learned by learners are extracted and the specific knowledge objects in the knowledge map are mapped.(3)Based on the model,a learning path recommendation strategy based on the inclusion relationship of knowledge map is proposed to help learners effectively learn target knowledge point objects and further solve the problem of lost and oversaturated in online learning.