Feature selection for unsupervised and supervised inference: The emergence of sparsity in a weight-based approach

Mathematics

• Feature Selection 100%
• Sparsity 78%
• Process Optimization 67%
• Ranking 50%
• Relevance 48%
• Least Squares 44%
• Spam 43%
• Text Classification 42%
• Text Processing 42%
• Electronic Mail 36%
• Bioinformatics 30%
• Computer Vision 29%
• Sorting 28%
• Nonnegativity 27%
• Subset 27%
• High Accuracy 25%
• Spectral Properties 24%
• Biased 23%
• Count 21%
• Branch 21%
• Evidence 20%
• Converge 15%

Engineering & Materials Science

• Feature extraction 57%
• Text processing 30%
• Electronic mail 27%
• Bioinformatics 27%
• World Wide Web 24%
• Sorting 24%
• Websites 21%
• Computer vision 20%