基于数据挖掘的环境污染档案数据统计系统设计研究

Design of Environmental Pollution Archives Data Statistics System based on Ddata Mining

针对目前中国自然环境污染档案数据统计效率较差的情况,基于数据挖掘技术,设计了新型环境污染数据统计系统。以射频和基带芯片为核心,利用大数据链路构建数据挖掘处理器,挖掘并缓存原始自然环境污染数据,设计系统自组织竞争网络,录入缓存数据并使用RBF神经网络对初始数据样本进行训练,将数据进行归一化处理,剔除原始环境数据中的错误数据,保证数据读入,将读入后的数据进行回归拟合和分布拟合并进行优化检验,确定函数特征,引入影响参数,分析环境数据的特征趋势,实现环境污染数据统计。

Aiming at the low statistical efficiency of natural environmental pollution archives data in China, a new statistical system of environmental pollution data was designed based on data mining technology. With rf and baseband chip as the core, the use of big data link processor to build data mining, mining and cache the original natural environment data, self-organizing competitive network design system, inputting the cached data, using RBF neural network to the initial training data sample, data normalization processing, eliminating errors in the data, the original environment data to ensure data read, to read in after the data regression fitting and distribution of fitting, and optimize test and determine the function characteristic, introducing the influence parameters, analysis of the characteristics of the environmental data, to complete the system design and implementation environment pollution statistics.