Dimensionality Reduction by Weighted Connections between Neighborhoods

Abstract

Dimensionality reduction is the transformation of high-dimensional data into a meaningful representation of reduced dimensionality. This paper introduces a dimensionality reduction technique by weighted connections between neighborhoods to improveK-Isomap method, attempting to preserve perfectly the relationships between neighborhoods in the process of dimensionality reduction. The validity of the proposal is tested by three typical examples which are widely employed in the algorithms based on manifold. The experimental results show that the local topology nature of dataset is preserved well while transforming dataset in high-dimensional space into a new dataset in low-dimensionality by the proposed method.

A Robust Stochastic Approach to Mineral Hyperspectral Analysis for Geometallurgy

“…A very interesting perspective is related to grouping or clustering surface zones to be labeled by expert knowledge and then applying regression or classification algorithms that help geological mapping at different scales (face mapping, drill core mapping). In this context there are some filters for image processing that are capable of smoothing content while preserving transitions [51]. For example, bilateral filtering is a simple, non-iterative scheme for edge-preserving smoothing.…”

Projection error evaluation for large multidimensional data sets

“…Dimensionality reduction deals with one special setting: a set of high-dimensional data points is mapped into low dimensionality such that data structure is preserved as much as possible [6]. Dimensionality reduction methods can also be thought of as a principled way to understand high-dimensional data [21]. They extract essential information from the high-dimensional data by mapping points from m-dimensional space to a d-dimensional space (d < m).…”