摘要
为了提高大规模数据挖掘效率和精度并降低算法复杂度,提出了一种基于人脑信息群智感知和机会跨层的认知数据挖掘机制。基于人脑神经元的激活原理训练感知模块,建立了机会认知的类脑群智架构。在堆栈和缓冲队列窗口之间建立映射关系,提出了类脑群智的机会认知数据挖掘进程。仿真实验结果表明,类脑群智的机会认知数据挖掘机制的数据特征值分布均匀。同时,与多层数据挖掘机制相比,运行时间短,数据精度高且平均聚类质量高。
In order to improve the efficiency and precision of large scale data mining and to reduce the complexity of the algorithm,a cognitive crowd data mining mechanism based on the knowledge of human brain is proposed.First of all,based on the principle of neural activation of human brain training module,the opportunistic cognitive model of brain was proposed.Then,a mapping relationship is established between the stack and the buffer queue window,and the data mining process is proposed.The simulation results show that the data characteristic value distribution of the cognitive data mining mechanism of the brain swarm intelligence is even.At the same time,compared with the multi-layer data mining mechanism,the running time is short,the data precision is high,and the average clustering quality is high.
出处
《计算机工程与设计》
北大核心
2017年第7期1828-1832,1871,共6页
Computer Engineering and Design
基金
四川省教育厅职成教科研课题(2015-2016年度)基金项目(职成教学[2015]10号)
关键词
大数据
数据挖掘
类脑群智
机会认知
跨层信息处理
big data
data mining
brain intelligent
opportunities cognition
cross layer information processing