“…Experiments show its advantages over some other methods. In the future, we will extend the proposed algorithm to various applications, such as computational mechanic [8], [9], [10], [11], [12], [13], multimedia [14], [15], [16], [17], [18], [19], [20], [21], [22], medical imaging [23], [24], [25], [26], [27], [28], [29], [30], [31], bioinformatics [32], [33], [34], [35], material science [36], [37], [38], high-performance computing [39], [40], [41], [42], [43], malicious websites detection [44], [45], [46], [47], biometrics [48], [49], [50], [51], etc. We will also consider using some other models to represent and construction the classifier, such as Bayesian network [52], [53],…”