Intelligent Courses Recommendation System of Collaborative Filtering Algorithm Based on K-means Clustering under Spark Platform

Abstract

Students in Chinese universities face a variety of electives. Their choice of the course impacts their scope of knowledge and their grade point, which influences their career or further learning. But in most condition, the most appropriate option will not be chosen due to the lack of detail and reference. Intelligent recommendation system was applied in multiple fields, collaborative filtering algorithm is one of common recommendation algorithm and is widely used. However, a collaborative filtering algorithm has the drawback in accuracy and user similarity. In that case, an improved algorithm based on K-means clustering is applied. The improved algorithm uses Dichotomous K-means algorithm to deal with the long distance of cluster centroid. As original similarity algorithm leads to the result of low user similarity, a mixed algorithm is applied. Traditional methodology of data computing and data storing cannot handle the demand of that much data set. As a big data platform, Spark has become a popular solution to these problems. University management combined with big data is a tendency[4].