基于K-means聚类和关联映射算法的数字画像匹配技术OA

A digital portrait matching technique based on K-means clustering and association mapping al-gorithm is proposed.The K-means clustering algorithm is applied to formulate the clustering partition strat-egy,and a large number of users' digital portraits are divided iteratively to form multiple portrait subsets.Combining latent factor models and matrix factorization concepts,the study generates corresponding cluster labels for each portrait subset.Using the association mapping algorithm,the paper analyzes the overall as-sociation between all cluster labels and the given digital portrait matching requirements in order to find the optimal portrait matching result.The experimental results show that the accuracy value of the digital por-trait matching results generated by this technology is not less than 0.9,indicating the superiority of the digital portrait matching results.