基于神经网络自适应搜索的大数据挖掘算法

Large Data Mining Algorithm Based on Neural Network Adaptive Search

大数据挖掘是实现数据库访问和数据库的关键信息特征提取的关键,传统的大数据挖掘算法采用粒子群进化方法,当数据特征空间中的干扰因素较多时,数据挖掘性能不好。提出一种基于神经网络自适应搜索的大数据挖掘算法,首先构建了大数据挖掘的神经网络训练模型,通过改进的BP神经网络训练进行数据特征提取和聚类处理,结合自适应搜索迭代方法进行大数据挖掘算法改进,提高数据挖掘过程中的聚类和特征提取性能。仿真结果表明,该大数据挖掘算法具有较高的特征匹配精度,挖掘准确性较高,自适应收敛性能较好,展示了较好的应用价值。

The data mining is ctitical to database access and key information feature extraction. In traditional data nfining algorithm using particle swarm optimization, when the data in the feature space have too many interference factors, data mining peffomlance degrades. This paper puts forward a neural network adaptive search based on large data mining algorithm. Firsdy, the neural network training model for large data mining is built, and the data feature extraction and data clustering processing are implemented by using improved BP neural network, and the large data mining algorithm is inaproved by using adaptive search iteration method to promote the data clustering and feature extraction during data mining. The simulation results show that this algorithm has high feature matching accuracy, high mining accuracy, better adaptive convergence performance and good application value.