O. Kouropteva, O. Okun, and M. Pietikäinen (Finland)
Data treatment and visualization, dimensionality reduction,machine learning.
An approach to visualization of high-dimensional data is proposed in this paper. It represents a visualization frame work that is based on the combination of four compo nents, namely, metric learning, intrinsic dimensionality es timation, feature extraction and feature selection. Thus, we view the visualization of high-dimensional data as a more complex process than the conventional dimension ality reduction. Though many successful applications of dimensionality reduction techniques for visualization are known, we believe that the sophisticated nature of high dimensional data often needs a concentration of several ma chine learning methods to solve the task, which is provided by the framework in experiments with real-world data.
Important Links:
Go Back
