A semi-parallel framework for greedy information-theoretic feature selection

Mathematics

• Feature Selection 82%
• Framework 35%
• High Performance Computing 11%
• Overfitting 10%
• Curse of Dimensionality 9%
• Large Data Sets 9%
• Preprocessing 8%
• Speedup 8%
• Paradigm 7%
• Computational Complexity 7%
• Filter 6%
• Optimization 5%
• Subset 4%

Business & Economics

• Feature Selection 100%
• Overfitting 10%
• Curse of Dimensionality 10%
• Computational Complexity 9%
• Filter 8%
• Big Data 7%
• High Performance 7%
• Paradigm 5%

Engineering & Materials Science

• Feature extraction 47%
• Large dataset 8%
• Big data 6%
• Computational complexity 6%
• Set theory 4%