Abstract:

Aiming at the problems of long mining time， poor mining effect and low mining performance in data mining due to the failure to extract relevant data features before data mining， a fast outlier mining algorithm in uncertain data set based on spectral clustering is proposed. The algorithm calculates the similarity of data according to unequal length sequences， and uses partial least square method to extract the features of uncertain data sets； Then， the data features are calculated based on spectral clustering algorithm to obtain the outlier index of the data； Finally， outlier mining of uncertain data sets is completed by outlier index. The experimental results show that the algorithm has the advantages of short mining time， good mining effect and high mining performance.

Key words: spectral clustering algorithm, uncertain data set, data outliers, fast mining, partial least squares method