Table 2 Overview of major feature extraction algorithms and their characteristics
From: A survey of dimension reduction and classification methods for RNA-Seq data on malaria vector
Feature extraction algorithms | Algorithms | Characteristics | Benefits and limitations |
|---|---|---|---|
Unsupervised Learning Approach | Principal Component Analysis [50] | Selects the most important genes and identifies transcriptional programs by extracting groups of genes that covary across a set of samples | Values taken by each variable do not all have the same importance and where the data may be contaminated with noise and contain outliers |
Supervised Learning Approach | New variables are confined in the rows of S, to wit, the variables observed are linearly collected independent components | Blind separation of independent sources from their linear combination | |
Partial Least Square (PLS) [53] | It is determined by a small number of latent characteristics It goes for discovering uncorrelated linear transformation of the initial indicator characteristics which have high covariance with the reaction characteristics | Latent components, PLS predicts reaction characteristics y, the assignment of regression, and reproduce initial matrix X, the undertaking of data modelling To optimize the covariance among the variable y and the initial predictor variables |