MACOC: A Medoid-Based ACO Clustering Algorithm

摘要: The application of ACO-based algorithms in data mining is growing over the last few years and several supervised unsupervised learning have been developed using this bio-inspired approach. Most recent works concerning focused on clustering, showing great potential techniques. This work presents an clustering algorithm inspired by ACO Clustering (ACOC) algorithm. proposed approach restructures ACOC from a centroid-based technique to medoid-based technique, where properties search space are not necessarily known. Instead, it only relies information about distances amongst data. new algorithm, called MACOC, has compared against well-known (K-means Partition Around Medoids) with ACOC. experiments measure accuracy for both synthetic datasets real-world extracted UCI Machine Learning Repository.