Research On Movie Recommendation Algorithm Of Association Rule Based On Combinatorial Similarity Clustering

With the improvement of people’s living standards,watching movies has become a leisure,entertainment,and lifestyle that our people are very keen on.This article intends to study an effective movie recommendation method,which can help users quickly find movies they like,and at the same time recommend movies to more suitable users,and play a positive role in the promotion of movies.The main research work and results of this paper are as follows:1.This paper proposes a new similarity calculation method—combined similarity.The similarity combines the advantages and disadvantages of modified cosine similarity,Jaccard similarity,and rating habit similarity according to different data sets according to a specific ratio,so that the similarity between users can be more accurately measured.The experimental results show that in the movie recommendation scene,the combined similarity is indeed better than the single similarity.2.This article improves the traditional K-means clustering algorithm.The traditional K-means algorithm uses Euclidean distance to measure the distance between data points,but Euclidean distance is not suitable for movie recommendation scenes.Therefore,this study uses combination similarity to replace Euclidean distance when measuring the distance of data points,and uses comparative experiments to prove that the recommendation effect of combination similarity based on the optimal weight ratio is better than the effect of Euclidean distance and each single similarity..3.Research and propose an association rule movie recommendation algorithm based on combined similarity clustering.The traditional association rule movie recommendation algorithm directly mines all data association rules,and then recommends movies according to the mined association rules.In this way,too much irrelevant data leads to low mining efficiency,and too few final rules lead to inaccurate recommendations.In this regard,this paper improves the method.Before mining association rules,first cluster users based on combination similarity,and then perform association rule mining in each category.When recommending a movie to a user,the category is determined first,and then the recommendation is made according to the association rules of the category to which the user belongs.Experiments have proved that the improved algorithm is indeed better than the traditional algorithm.The research contributions of this paper are: based on a single similarity,a combined similarity is proposed;the combined similarity is used to replace the Euclidean distance in the K-means algorithm;the improved K-means algorithm is used to improve the traditional association rule recommendation.Experiments show that combined similarity is better than Euclidean distance and single similarity,and the association rule movie recommendation algorithm based on combined similarity clustering is better than the traditional association rule recommendation algorithm.The optimization algorithm proposed in this paper is feasible.