A novel genetic clustering algorithm with variable-length chromosome representation
Ming-an Zhang, Yong Sheng Deng, Dongxia Chang · 2014
The paper proposed a new genetic clustering algorithm with variable-length chromosome representation (GCVCR), which can automatically evolve and find the optimal number of clusters as well as proper cluster centers of the data set. A new clustering criterion based on message passing between data points and the candidate centers described by the chromosome are presented to make the clustering problem more effective. The simulation results show the effectiveness of the proposed algorithm.