基于仿生算法的显著性特征数据挖掘方法

摘要

对大数据信息库进行显著性特征数据挖掘,在故障诊断和模式识别等领域具有重要的应用价值.针对传统方法采用大量分布式计算方法对显著性特征数据进行挖掘时,容易出现局部收敛,计算开销较大,数据挖掘实时性不好等问题,提出基于粒子群仿生算法的大数据信息库显著性特征数据挖掘模型.通过提取显著性特征数据的数据结构特征,完成显著性特征数据信息流拟合分析.采用粒子群仿生算法对显著性特征数据的挖掘过程进行跟踪训练,实现挖掘误差修正,完成显著性特征数据挖掘.仿真结果表明,采用该算法进行显著性特征数据挖掘的全局寻优和收敛性较好,挖掘精度较高.%The significant feature data mining of large data information database has important application value in the field of fault diagnosis and pattern recognition.The traditional method uses a large number of distributed computing method to mine the significant characteristic data,which has local convergence,high computational cost,inferior real-time performance of data mining.Data mining model of big data information significant characteristics based on the particle swarm bionic algorithm is proposed.By extracting data structure characteristics of data with significant features,its information flow fit analysis is made.Using particle swarm bionic algorithm to track and train its mining process,the mining error correction,and the significant data mining are achieved.The simulation results show that the algorithm has better global optimization,convergence,and higher precision.