Big Data Modeling of Multi-interval Nested Correlation Function for Industrial Structure Optimization in Hainan

Abstract

Big data is a concept driven by demand. With the popularity of the database system and the expansion of Internet services, the data available to enterprises or individuals are expanding, and it is highly difficult to meet the data analysis needs with the existing technologies in the big data era. Therefore, it is necessary to explore new theories and methods to support the application of multi-interval nested correlation function big data. Although the 4V attribute of the big data modeling of multi-interval nested correlation function for industrial structure optimization in Hainan (ISOH) has been widely discussed, most of them are still described as the representation of multi-interval nested correlation function big data. Hence, it is very difficult to abstract a unified data format, and it is necessary to further identify the technical features that can be used for data formatting. For the big data application demand of multi-interval nested correlation function with the industrial structure optimization as the main technical feature, the industrial structure optimization in Hainan is adopted as the data expression carrier in the pape. On this basis, the corresponding multi-interval nested correlation function big data classification model and mining operator are designed. At the same time, the algorithm corresponding to the key steps is constructed for the key problems to be solved in the classification mining of multi-interval nested correlation function big data. The rationality of the micro-cluster merging technology and sample data reconstruction method provided in the paper is verified theoretically. The experiments show that the big data algorithm of multi-interval nested correlation function based on the industrial structure optimization in Hainan put forward in this paper can not only greatly reduce the communication cost between network nodes but also achieve an average improvement of about 10% in global mining precision (compared to the existing typical algorithm DS-means). Although the time spent is slightly higher than DS-means, the difference between them is minimal under different data capacity tests, and the trend of time rising is similar.