Doubly supervised embedding based on class labels and intrinsic clusters for high-dimensional data visualization

Keyphrases

• Dimensionality Reduction 100%
• Class Label 100%
• Natural Grouping 100%
• Unsupervised Dimensionality Reduction 100%
• Supervised Embedding 100%
• High-dimensional Data Visualization 100%
• Class Separability 66%
• Overall Structure 33%
• State-of-the-art Techniques 33%
• Feature Space 33%
• Label Information 33%
• Class Information 33%
• Unsupervised Learning 33%
• Decision-making Process 33%
• High-dimensional Data 33%
• Dimensional Relationships 33%
• Data Visualization 33%
• Cluster Structure 33%
• Significant Distortions 33%
• Main Principle 33%
• Supervised Methods 33%
• T-distributed Stochastic Neighbor Embedding (t-SNE) 33%
• Reduced Features 33%

Computer Science

• High Dimensional Data 100%
• Scientific Visualization 100%
• Decision-Making 50%
• Feature Space 50%
• Supervised Method 50%
• Class Information 50%
• Dimension Reduction Method 50%
• Cluster Structure 50%