Research on the Application of Improved Collaborative Filtering Algorithm in Course Recommendation
Qing Li, Yingying Wen, Wenwen Zhou, Huanhuan Wan, Aiguo Zhang, Mingxuan Liu, Luqian Xu · 2023
In the era of big data, the amount of resources on educational platforms is constantly increasing, making it difficult for many learners to choose the courses they really need from a massive number of online courses. Collaborative filtering algorithms are widely used in recommendation systems. Traditional collaborative filtering algorithms usually use the cosine similarity formula for interest similarity calculation, but in many cases, popular courses can affect the recommendation results and cannot reflect student needs well. This article proposes an improved solution to the cosine similarity calculation formula of traditional collaborative filtering algorithms, which can effectively suppress the influence of popular courses on the actual similarity of students. Experimental results show that this method can efficiently recommend online courses on educational platforms.