A Bayesian Classification Intrusion Detection Method Based on the Fusion of PCA and LDA
Algorithm 1
(1)
Remove the average value and subtract every element in matrix X from , is the mean value of the dimension i of the sample matrix (matrix column element), .
(2)
Computing covariance matrix C of matrix X, .
(3)
Calculate the eigenvalues and eigenvectors of the covariance matrix C.
(4)
Sort them according to the numerical value of the eigenvalues, retaining the first k eigenvectors.
(5)
It is generally believed that the first three eigenvectors are the most representative of the overall information. When the data are affected, the three eigenvectors are most susceptible to contamination. If weighting is performed, the degree of influence can be reduced, the accuracy can be improved, and the data can be converted to the above. A new matrix of k eigenvectors and is defined as in (7).
