Feature Learning for Nonlinear Dimensionality Reduction toward Maximal Extraction of Hidden Patterns

TL;DR

This paper introduces FEALM, a feature learning framework that enhances nonlinear dimensionality reduction by generating optimized projections to better reveal hidden patterns in high-dimensional data.

Contribution

FEALM is a novel framework that optimizes data projections for nonlinear DR, incorporating a new graph dissimilarity measure and interactive visualization tools.

Findings

FEALM effectively captures hidden patterns in synthetic datasets.

FEALM outperforms existing methods in revealing intrinsic data structures.

Interactive visualizations aid in interpreting DR results.

Abstract

Dimensionality reduction (DR) plays a vital role in the visual analysis of high-dimensional data. One main aim of DR is to reveal hidden patterns that lie on intrinsic low-dimensional manifolds. However, DR often overlooks important patterns when the manifolds are distorted or masked by certain influential data attributes. This paper presents a feature learning framework, FEALM, designed to generate a set of optimized data projections for nonlinear DR in order to capture important patterns in the hidden manifolds. These projections produce maximally different nearest-neighbor graphs so that resultant DR outcomes are significantly different. To achieve such a capability, we design an optimization algorithm as well as introduce a new graph dissimilarity measure, named neighbor-shape dissimilarity. Additionally, we develop interactive visualizations to assist comparison of obtained DR results and interpretation of each DR result. We demonstrate FEALM's effectiveness through experiments and case studies using synthetic and real-world datasets.