Research On Algorithm Of Course Recommendation Based On MOOC

Data mining is a technology that analyzes the relationship,trend and mode of data by analyzing massive data.It is a technology that integrates artificial intelligence,database technology,pattern recognition,machine learning,data visualization and statistics.Technology of cross-disciplinary.Clustering,regression analysis and classification,deviation analysis,and association rules are the common methods of data mining modeling.The algorithms include artificial neural network,decision tree,genetic algorithm,neighbor algorithm,rough set method,fuzzy Set method,that what called statistical analysis function and thros derivation.In this paper,we mainly use clustering,classification and correlation analysis and related algorithms to study.(MOOC)technology in the online network education has been rapid popularization and development,this article to a university "cloud classroom" MOOC platform as the basis,the use of correlation analysis,collaborative filtering technology,the use of large-scale online courses,Which relaye analysis for the timing of the student achievement and the intrinsic relevance,the study of the students' pre-elective course alway select that specific course to those students.The study of the situation to predict,and analyze the relationship between the results of the course,trying to find the current college students to learn less efficient,lack of motivation,low rate of knowledge conversion program,and strive to recommend more suitable for their students,there are Conducive to its long-term development and planning courses.This paper analyzes the principles and methods of association rules and collaborative filtering.According to the actual business scenarios,the analysis and design of the response system to the recommendation system is put forward,and the proposed scheme is put forward.Then,the relationship between the students' learning situation in the Muji system is analyzed by the correlation analysis The data are collected and sorted,and the students 'achievement data are merged and integrated to obtain the time series data of the students' elective courses and achievements.The data is preprocessed by discretization and data thinning,which can be highly structured and can be used Weka(An open source intelligent analytics platform developed by the University of New Zealand at Waikato)to deal with the data entry.Then,using the association rule analysis algorithm Apriori to correlate the student achievement data,the relationship between the student's historical achievement and the course is obtained by using the correlation between the different courses based on finding and finding the same specialty.In order to get in a pre-course to obtain excellent results in the case,and then learn what other courses are more likely to achieve better results.Among them,we use student-based and curriculum-based togther filtering technology,according to the similarity between students,similarity between the curriculum analysis and weighted sort,remove unreasonable,illogical and redundant data,to meet the requirements of the recommended results The Finally,a joint recommendation is come from what the results showed rely the association analysis and collaborative filtering recommendation.In the recommended design of this study,the students' "talent" and "interest" are incorporated into the conditional factors of the recommended algorithm.As a recommendation algorithm,in the electricity business,more consideration is the user's shopping habits,income conditions and other factors;as a curriculum,the school is more consideration is the relationship between the curriculum,as well as the comprehensive effectiveness to the students to cultivate.But the actual situation is that each student has their own interest in the direction of their own good subjects,individualized teaching will be more conducive to explore the potential of students to help students grow.