Abstract

The article describes a new unique method of multidimensional data visualization. It has been developed as modified observational tunnels method, which was previously known and used many times. The modification consists in supplementing the observational tunnels method used for visualization of multidimensional data with the concept of perspective. In this way, the orientation and navigation in multidimensional space are largely facilitated. The differences in effects of observational tunnels method and perspective-based observational tunnels method have been presented. The effectiveness of the new visualization method has been compared with selected four well-known methods of multidimensional data visualization: parallel coordinates, orthogonal projection, principal component analysis, and multidimensional scaling. The research revealed that the perspective-based observational tunnels method sometimes makes it possible to obtain information about significant features of analyzed data even when other methods selected for comparative studies are not able to show it. This article includes a presentation of the views of 5-dimensional data obtained from the print recognition process, which allowed the author to state that the features chosen for the development of spatial features are, in this case, sufficient for the correct recognition process. The previously published ranking presenting seven different methods of multidimensional data visualization was supplemented with the perspective-based observational tunnels method. This ranking was conducted using 7-dimensional data describing different types of coal. Thus, it was shown that, in this case, the presented method constitutes the efficient tool among other qualitative visualization analysis methods.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call