Packaging Big Data Visualization Based on Computational Intelligence Information Design

Abstract

A method based on a computational intelligence information model is proposed to study the visualization of large data packages. Since the CAIM algorithm only considers the distribution of the largest number of classes in an interval, it offers an optimization method and simultaneously determines the appropriate stopping conditions to avoid overcrowding. The effectiveness of the improved algorithm has been experimentally proven. Methods of character reduction and weight determination are used to reduce the index and weight, establishing a large packaging information system. Experimental results show that the improved algorithm in this article produces more classification rules than the CAIM algorithm, because the discrete intervals created by the CAIM algorithm are relatively simple, but the classification rules are few, but less than the number of CAIM algorithms. Classification rules are generated by entropy-based sampling algorithms. This can make the classification rules simple and universal, and it is clear that the optimal sampling algorithm is more accurate than the CAIM algorithm.