摘要
提出基于朴素贝叶斯的大数据模糊随机挖掘仿真方法,为用户挖掘海量数据特征并从中发现可用数据提供有效途径。该方法依据数据间的关联规则,对具备非线性特征的大数据进行融合处理,利用模糊层次聚类算法依据融合后大数据获取大数据语义关联特征;将语义关联特征作为朴素贝叶斯分类器的输入,输出大数据模糊随机挖掘结果。仿真结果表明,上述方法融合大数据时的关联规则支持度最大为100%,大数据融合效果较好;在大数据量为100GB时,其提取大数据语义关联特征时的概率化特征条件引入量高达96%;模糊随机挖掘大数据时,大数据空间聚焦能力较好,可有效实现大数据模糊随机挖掘。
A simulation method of fuzzy random mining for big data based on naive Bayes is proposed,which pro-vides an effective way for users to mine massive data features and find available data.According to the association rules between the data,the method fuses the big data with nonlinear characteristics,and uses the fuzzy hierarchical clustering algorithm to obtain the semantic association features of the data based on the fused big data.The semantic association feature is used as the input of naive Bayes classifier to output the fuzzy random mining results of big data.The simulation results show that the maximum support degree of association rules is 100%,and the fusion effect of big data is good.When the volume of big data is 100GB,the introduction of probabilistic feature conditions in extracting semantic association features of big data is as high as 96%.When mining big data with fuzzy randomness,the ability of focusing on big data space is better,which can effectively realize fuzzy random mining of big data.
作者
陈晓姗
张国华
CHEN Xiao-shan;ZHANG Guo-hua(Taizhou College,Nanjing Normal University,Taizhou Jiangsu 225300,China)
出处
《计算机仿真》
北大核心
2023年第11期428-432,共5页
Computer Simulation
基金
2020年教育部产学研育人项目(202002320023)。
