Heterogeneous High Performance Data Mining System for Intelligent Data

Abstract

In order to improve the utilization rate of internet data under heterogeneous distribution, increase the diversified usage functions and data transmission rate of the internet, and reduce the running time of the internet, it is necessary to mine internet data under heterogeneous distribution. The author proposes an ontology based optimization method for internet data mining under heterogeneous distribution; This method first preprocesses and selects data features from internet data under heterogeneous distribution, and uses a feature selection decision system to select features from the mining data. Based on this, information entropy is used to filter internet data under heterogeneous distribution. During the filtering process, the theoretical values filtered by information entropy are reduced to obtain the optimal data filtering value, finally, based on the various data information obtained in the preprocessing, the iterative calculation results of the information gain value in the decision tree generation algorithm are used to high-precision mine internet data under heterogeneous distribution; The simulation experimental results demonstrate that theproposed method improves the flexibility of internet data operations under heterogeneous distribution, increases the recyclability of internet data, and makes internet operations under heterogeneous distribution more concise and efficient, providing a strong basis for research and development in this field.