Abstract | ||
---|---|---|
This paper presents a new learning strategy for the clustering algorithms based on Self-Organizing Map. Our contribution relies on the competitive phase of this unsupervised learning algorithm and proposes a new strategy for choosing the most active cell/neuron. This new strategy is to choose the most active neuron taking into account its historical activations, learned in a voting matrix from the dataset. Indeed, the use of this historic neighbourhood, allows introducing of topological constraints in the final geometry of the map. This new unsupervised learning approach allows discovering of data structure with a better quality (lower topographic error and better clustering purity). The proposed approach was validated on multiple datasets of different sizes and complexities, and the experimental validation shows promising results. |
Year | DOI | Venue |
---|---|---|
2009 | 10.1109/ICMLA.2009.21 | Miami Beach, FL |
Keywords | Field | DocType |
clustering purity,unsupervised learning algorithm,new competitive strategy,active neuron,new unsupervised learning approach,new strategy,clustering algorithm,self organizing map learning,better quality,active cell,new learning strategy,history,data mining,indexes,unsupervised learning,competitive strategy,data structure,memory,prototypes,self organizing map,clustering algorithms,clustering,active learning,machine learning | Data structure,Competitive learning,Pattern recognition,Voting,Computer science,Topographic map,Competitive advantage,Self-organizing map,Unsupervised learning,Artificial intelligence,Cluster analysis,Machine learning | Conference |
ISBN | Citations | PageRank |
978-0-7695-3926-3 | 0 | 0.34 |
References | Authors | |
4 | 2 |
Name | Order | Citations | PageRank |
---|---|---|---|
Nistor Grozavu | 1 | 67 | 16.76 |
Younès Bennani | 2 | 269 | 53.18 |