Research Article
An Approach for Demand Forecasting in Steel Industries Using Ensemble Learning
Algorithm
1 Steps for the implementation of principal component analysis (PCA).
| Input: dimensional input data matrix with number of samples , and variance threshold | | Output: reduced dimensional data matrix , | | Load , and calculate mean for each feature, for subtract the mean from each corresponding dimension, for and | | / Make each signal uncorrelated to each other / | | Calculate covariance matrix of | | Solve the as , where is the matrix of eigenvector and is the diagonal matrix containing eigenvalues on both sides of the diagonal matrix | | Sort the eigenvector matrix in the descending order to the first eigenvector that have variance and form a projection matrix | | Finally, project on the PCA space, |
|