Research on Network Aware Data Fusion Algorithm Based on Fuzzy Time Series

 Jia Chen, Yingxue Cai, Song Hu, Sibo Huang, Hui Hu, Zhaoquan Cai*

Huizhou University, Huizhou, 516000, China

Abstract: In the environment of big data, the data mining and fusion of network awareness is affected by the interference between the small disturbances of the data, which leads to the poor clustering characteristics of the information fusion. A network aware data fusion method based on improved fuzzy time series analysis algorithm is proposed. The data stream of web address is reorganized in the distributed structure, and the phase space reconstruction technique is used to deal with the autocorrelation feature matching of the fuzzy time series attractor ｌｏａｄ. Combined with association rule fuzzy pairing method, fuzzy time series analysis is used to realize the optimal clustering of network perceptual data. According to the result of data clustering, the accurate fusion of network perceptual data is realized, and the feature compression of fused high-dimensional network perceptual data is carried out, so as to reduce computational overhead. The simulation results show that the proposed method has good clustering ability, it has strong ability to resist inter-class disturbance and strong ability of network perceptual data fusion.

Keywords: Big data; Fuzzy time series; Fusion; Feature clustering