Review of Dimension Reduction Methods

Abstract

Purpose: This study sought to review the characteristics, strengths, weaknesses variants, applications areas and data types applied on the various Dimension Reduction techniques. Methodology: The most commonly used databases employed to search for the papers were ScienceDirect, Scopus, Google Scholar, IEEE Xplore and Mendeley. An integrative review was used for the study where 341 papers were reviewed. Results: The linear techniques considered were Principal Component Analysis (PCA), Linear Discriminant Analysis (LDA), Singular Value Decomposition (SVD), Latent Semantic Analysis (LSA), Locality Preserving Projections (LPP), Independent Component Analysis (ICA) and Project Pursuit (PP). The non-linear techniques which were developed to work with applications that have complex non-linear structures considered were Kernel Principal Component Analysis (KPCA), Multi-dimensional Scaling (MDS), Isomap, Locally Linear Embedding (LLE), Self-Organizing Map (SOM), Latent Vector Quantization (LVQ), t-Stochastic neighbor embedding (t-SNE) and Uniform Manifold Approximation and Projection (UMAP). DR techniques can further be categorized into supervised, unsupervised and more recently semi-supervised learning methods. The supervised versions are the LDA and LVQ. All the other techniques are unsupervised. Supervised variants of PCA, LPP, KPCA and MDS have been developed. Supervised and semi-supervised variants of PP and t-SNE have also been developed and a semi supervised version of the LDA has been developed. Conclusion: The various application areas, strengths, weaknesses and variants of the DR tech

Multi-omics data integration approaches for precision oncology

“…46 t-Stochastic neighbor embedding (t-SNE) t-Stochastic Neighbor Embedding (t-SNE) is an unsupervised NLDRT which generates probability distributions from pairwise distances. 47 This technique relies on conditional probability to transform high dimensional data into a low-dimensional space 33 and it is a common tool for data visualization. 48 Nowadays, it is frequently used in the visualization of singlecell RNA-sequencing (scRNA-seq) data to assess cellular heterogeneity in tumor samples.…”

“…The main disadvantage of UMAP is the fact that it is a relatively new technique and therefore lacks maturity. 47 The role of dimensionality reduction in multi-omics data integration DRTs are essential to discover new clusters/subtypes 54 and identify outliers in individual omics data sets. 55 They play a key role in exploring the association of samples subtypes with outcome/survival.…”

“…49 Even though t-SNE is frequently used in single-cell analysis, a slow computation time and loss of large scale information are among its main limitations. 47 Therefore these drawbacks must be taken into account when using t-SNE for multi-omics data integration.…”

“…The main disadvantage of UMAP is the fact that it is a relatively new technique and therefore lacks maturity. 47…”

Classification of Different Shoulder Girdle Motions for Prosthesis Control Using a Time-Domain Feature Extraction Technique

“…A G affinity graph with labeled and unlabeled points was developed to examine the complexity of the underlying geometry and learn responses from given data. The embedding function is realized utilizing these responses and standard regression (Nanga, et al, 2021).…”