基于区块链与模糊聚类算法的区域大数据分析技术研究  

Research on regional big data analysis technology based on blockchain and fuzzy clustering algorithm

作　　者：何颖[1] HE Ying(Chengyi College,Jimei University,Xiamen 361000,China)

机构地区：[1]集美大学诚毅学院,福建厦门361000

出　　处：《现代电子技术》2025年第6期52-56,共5页Modern Electronics Technique

基　　金：福建省社科青年基金项目(FJ2016C142)。

摘　　要：金融数据具备非线性、高维度的特点,同时对安全性有较高的要求。文中结合区块链技术和模糊聚类算法,提出一种面向区域互联网金融的异常数据分析模型,该模型由异常数据分析算法和隐私保护算法组成。异常数据分析算法针对模糊均值聚类算法处理高维非线性数据能力弱的缺点,使用深度信念网络进行改进,进而提升模型的数据处理能力。隐私保护使用差分隐私保护算法,在不利用背景知识的前提下完成数据的保护,同时保证了数据的可用性。在实验测试中,将所提模糊聚类算法与常用的主流K-Means算法、DPC算法进行了对比,结果表明:所提算法的性能在所有对比算法中最优;与此同时,加入隐私保护算法后对聚类结果的影响保持在0.021以内,充分证明了该算法性能的优越性。Financial data has the characteristics of nonlinearity and high dimensionality,while also requiring high security.By combining with blockchain technology and fuzzy clustering algorithm,an anomaly data analysis model for regional internet finance is proposed.This model is composed of anomaly data analysis algorithm and privacy protection algorithm.In allusion to the weakness of the fuzzy mean clustering algorithm in processing high-dimensional nonlinear data,the anomaly data analysis algorithm can use the deep belief networks to improve the data processing capabilities of the model.In the privacy protection,the differential privacy protection algorithm can be used to protect data without utilizing background knowledge,while ensuring data availability.In the experimental testing,the proposed fuzzy clustering algorithm was compared with commonly used mainstream K-Means and DPC algorithms.The results show that the performance of the proposed algorithm is the best among all comparison algorithms.At the same time,the impact of incorporating privacy protection algorithm on clustering results can remain within 0.021,fully demonstrating the superiority of the performance of this algorithm.

关 键 词：模糊聚类算法 区块链技术 异常数据识别 深度信念网络 差分隐私保护算法 区域数据分析 

分 类 号：TN919.5-34[电子电信—通信与信息系统] TP391[电子电信—信息与通信工程]