异构式分布下的internet数据挖掘方法优化研究

为了提高异构式分布下的internet数据的利用率,增加internet的多样化使用功能和数据传输率,减少internet运行的时间,需要对异构式分布下的internet数据进行挖掘。当前的数据挖掘方法多是先采用SOM系统的可视化功能对异构式分布下的internet数据进行聚类,然后根据聚类结果的计算完成对异构式分布下的internet数据挖掘。但该方法存在操作过程复杂,internet数据经常性丢失的问题。为此,提出了一种基于本体论的异构式分布下的internet数据挖掘优化方法。该方法首先对异构式分布下的internet数据进行预处理选取出数据特征,并利用特征选择决策系统对挖掘数据进行特征选择,在此基础上利用信息熵实现异构式分布下的internet数据的过滤,过滤过程中通过信息熵数据过滤的理论值减小的变动,得到最佳数据过滤值,最后以预处理中获得的各项数据信息为基础,采用决策树生成算法中的信息增益值的迭代计算结果对异构式分布下的internet数据进行高精度挖掘。仿真实验结果证明,所提方法提高了异构式分布下的internet数据操作的灵活度,增加了internet数据的可循环利用率,使异构式分布下的internet操作更加简洁化、高效率化,为该领域的研究发展提供了强有力的依据。

Under the heterogeneous distribution of Internet data mining method optimization research

In order to improve the utilization rate of the Internet data under the heterogeneous type distribution, increase the diversification of the Internet use function and data transfer rate, reduce the operation of the Internet time, needs to be under the heterogeneous distribution of Internet data mining. The current data mining method is to adopt the visual function of SOM system under the heterogeneous distribution of Internet data clustering, then according to the clustering results of calculation with heterogeneous distribution of Internet data mining. But this method is a complex process of operation, regular Internet data missing problem. To this end, this paper proposes a heterogeneous distribution based on ontology of Internet data mining method. The method of first selection under the heterogeneous distribution of Internet data preprocessing the data characteristics, and use the feature selection decision-making system for mining data feature selection, on the basis of using information entropy under the heterogeneous distribution of Internet data filter, filter by information entropy in the process of data, the change of the theoretical value of reducing data filtering to get the best value, and finally on the basis of the pretreatment of the data obtained from the information, using the decision tree generation algorithm of iterative calculation results of information gain value to high precision under the heterogeneous distribution of Internet data mining. Simulation experimental results show that the proposed method improves the heterogeneous distribution of Internet data operation flexibility, increased the recycled utilization rate of Internet data, makes the heterogeneous distribution of the Internet more concise, efficient operation, the research in the field development provides a strong basis.