Combinación de clusterizadores difusos mediante voto posicional para clustering robusto de documentos

The combination of multiple clustering processes provides a means for building robust document clustering systems. This work focuses on the consolidation of fuzzy clusterings, proposing two consensus functions for soft cluster ensembles based on the Borda and Condorcet positional voting strategies. Experiments conducted on two document corpora reveal that the proposed soft consensus functions are capable of yielding consensus partitions of comparable or superior quality to those obtained by state-of-the-art clustering combiners, although their computational complexity is higher due to the ranking process inherent to positional voting techniques.