Improved CURE algorithm and application of clustering for large-scale data

Aiming at the classification problem of large-scale document information, a large-scale data clustering algorithm based on improved CURE algorithm is proposed. By clustering the data partition and the initial class of after partition, data tracking, the large-scale data hierarchical clustering and sample classification is achieved, that better solved the balance of clustering quality and clustering effectiveness. Taking the actual document processing of Large-scale network data, the experiment results show that the algorithm is efficient.