Skip to main content

Advertisement

Fig. 4 | Big Data Analytics

Fig. 4

From: Identification of disease-distinct complex biomarker patterns by means of unsupervised machine-learning using an interactive R toolbox (Umatrix)

Fig. 4

Screenshots of interface components “iUmapIsland” and “iClassification” of the interactive “Umatrix” R library. a: The “iUmapIsland” component provides an interface to manually cut a so-called “island” out of the toroid Umatrix. The island should contain each part of the U-matrix only once. It is usually cut along cluster borders enhancing the visual data structure of emerging in the U-matrix. The following user interactions are implemented: ❶ the trained U-matrix is shown at the right of the interface panel. Using the mouse, on the toroid U-matrix (see Fig. 3) a unique region can be marked that contains every best matching unit (BMU) only once. ❷ The marked region can be cut out of the toroid U-matrix providing the so-called “island” as the standard representation of the U-matrix. ❸ A prior classification can be loaded from a structured text file. ❹ The display of the U-matrix can be visually modified such as changing the size, the diameter of the best matching units. ❺ The numerical results can be saved to a file and the interface is finally closed. b: The “iClassification” component is an interactive shiny tool that visualizes a given U-matrix and allows the user to select areas and mark them as clusters. The following user interactions are implemented: On the island, ❻ areas located in the same regions can be marked with the mouse as clusters, ❼ that can be added to the available clauses, or, clusters can be deleted manually. ❹ The display of the U-matrix can be visually modified such as changing the size, the diameter of the best matching units or ❾ making the colors slightly transparent which enhanced data structure visibility. ❽ Selection of the projection grid as either toroid where opposite edges are connected or planar. The figure has been created using the R software package (version 3.4.0 for Linux; http://CRAN.R-project.org/ [24]) using the R package “Umatrix” (https://cran.r-project.org/package=Umatrix)

Back to article page