Frequency-sensitive competitive learning for scalable balanced clustering on high-dimensional hyperspheres

Keyphrases

• K-means 100%
• Competitive Learning 100%
• Hypersphere 100%
• Balanced Clustering 100%
• Number of Clusters 66%
• Generative Models 66%
• Balanced Clusters 66%
• Batch Mode 33%
• Maximum Likelihood 33%
• Poor Performance 33%
• Number of Data 33%
• Three-frequency 33%
• Static Data 33%
• Curse of Dimensionality 33%
• Text Data 33%
• High-dimensional Data 33%
• Dimension Effect 33%
• Well-balancing 33%
• Input Vector 33%
• Expectation Maximization 33%
• Streaming Data 33%
• Text Document 33%
• High-dimensional Input 33%
• Cluster Center 33%
• High-dimensional Space 33%
• Document Clustering 33%
• Von Mises-Fisher Distribution 33%
• Competitive Learning Mechanism 33%
• Unbalanced Clusters 33%

Mathematics

• Clustering 100%
• K-Means 100%
• Hypersphere 100%
• Dimensional Space 66%
• Maximum Likelihood 33%
• Data Point 33%
• Curse of Dimensionality 33%
• Learning Mechanism 33%
• Unit Length 33%
• Dimensional Data 33%
• Cluster Center 33%
• Text Data 33%
• Input Vector 33%
• Text Document 33%
• Data Streaming 33%

Computer Science

• Generative Model 100%
• Experimental Result 50%
• maximum-likelihood 50%
• Learning Mechanism 50%
• Dimensional Space 50%
• Data Streaming 50%
• High Dimensional Data 50%
• Document Clustering 50%
• High Dimensional Space 50%