Location algorithm of fuzzy outliers in big data networks
| Autor: | Rentai Chen |
|---|---|
| Rok vydání: | 2021 |
| Předmět: |
Computer Networks and Communications
Computer science business.industry Anomaly (natural sciences) Node (networking) Big data Lattice phase equaliser Fuzzy logic ComputingMethodologies_PATTERNRECOGNITION Data point Outlier Electrical and Electronic Engineering Cook's distance business Algorithm Information Systems |
| Zdroj: | Wireless Networks. 28:2785-2793 |
| ISSN: | 1572-8196 1022-0038 |
| DOI: | 10.1007/s11276-021-02706-y |
| Popis: | The traditional Hadoop-based anomaly data recognition algorithm for big data networks does not suppress the disturbance components of the data attributes of anomalous nodes. A new algorithm for locating and identifying fuzzy anomaly data in big data network is proposed. In the big data network environment, adaptive cascade notch filter is used to eliminate data interference, and second-order lattice filter is used to locate abnormal node data. The parameters of the fuzzy linear regression model are estimated, and the fuzzy Cook distance is solved. The data points with the largest fuzzy Cook distance are regarded as fuzzy abnormal data, and the data location and recognition are realized. The experimental results show that the average recall rate of the proposed algorithm for locating fuzzy outlier data is 93%, and the locating probability of the proposed algorithm for fuzzy outlier data is 92% when the signal-to-noise ratio is −30 dB. The proposed algorithm can accurately identify the fuzzy outlier data in big data network by Cook distance, and has better locating and identifying effect for fuzzy outlier data in big data network. |
| Databáze: | OpenAIRE |
| Externí odkaz: |
|
Full text from SpringerLink