基于改进混沌分区算法的模糊信息抽取

在大数据环境下进行模糊信息挖掘抽取中受到数据之间的小扰动类间干扰的影响,导致信息抽取的特征聚类性不好。为此提出一种基于改进混沌分区算法的模糊信息抽取方法,对高维数据信息流进行分布式结构重组,以Lorenz混沌吸引子作为训练测试集进行大数据模糊信息抽取的自适应学习训练,采用相空间重构技术对大数据的混沌吸引子负载特征量进行自相关特征匹配处理,提取模糊信息的平均互信息特征量,结合关联规则模糊配对方法进行大数据混沌分区,实现模糊信息的优化聚类,根据数据聚类结果实现模糊信息准确抽取,对抽取的高维模糊信息进行特征压缩,降低计算开销。仿真结果表明,采用该方法进行大数据样本序列的模糊信息抽取的聚类性较好,抗类间扰动能力较强,模糊信息抽取的准确概率较高,在数据挖掘和特征提取中具有很好的应用价值。

Fuzzy information extraction based on improved chaotic partition algorithm

In the environment of big data,the interference between the small disturbances of the data affects the fuzzy information extraction,which leads to the poor clustering characteristics of information extraction.This paper proposed a fuzzy information extraction method based on the improved chaotic partition algorithm.It reorganized the high dimensional data information flow with distributed structure,and used the Lorenz chaotic attractor as the training test set for the adaptive learning training of big data fuzzy information extraction.It used the phase space reconstruction technique match big data’s chaotic attractor load with autocorrelation feature matching,and extracted the average mutual information feature quantity of fuzzy information.Through realizing the optimal clustering of fuzzy information,it realized the accurate extraction of fuzzy information according to the result of data clustering,carried out the feature compression of the extracted high-dimensional fuzzy information,and reduced the computational overhead.The simulation results show that,using this method to extract fuzzy information from big data sample sequence has good clustering property,strong ability to resist inter-class disturbance,and high accurate probability of fuzzy information extraction.It has a good application value in data mining and feature extraction.