Map/Reduce Affinity Propagation Clustering Algorithm
Wei-Chih Hung 1, Chun-Yen Chu 1, Yi-Leh Wu 1, and
Cheng-Yuan Tang 2
1. Department of Computer Science and Information Engineering, National Taiwan University of Science and Technology, Taipei, Taiwan
2. Department of Information Management, Huafan University, New Taipei City, Taiwan
2. Department of Information Management, Huafan University, New Taipei City, Taiwan
Abstract—The Affinity Propagation (AP) is a clustering algorithm that does not require pre-set K cluster numbers. We improve the original AP to Map/Reduce Affinity Propagation (MRAP) implemented in Hadoop, a distribute cloud environment. The architecture of MRAP is divided to multiple mappers and one reducer in Hadoop. In the experiments, we compare the clustering result of the proposed MRAP with the K-means method. The experiment results support that the proposed MRAP method has good performance in terms of accuracy and Davies–Bouldin index value. Also, by applying the proposed MRAP method can reduce the number of iterations before convergence for the K-means method irrespective to the data dimensions.
Index Terms—affinity propagation, map/reduce, hadoop, K-means, clustering algorithm
Cite: Wei-Chih Hung, Chun-Yen Chu, Yi-Leh Wu, and Cheng-Yuan Tang, "Map/Reduce Affinity Propagation Clustering Algorithm," International Journal of Electronics and Electrical Engineering, Vol. 3, No. 4, pp. 311-317, August 2015. doi: 10.12720/ijeee.3.4.311-317
Cite: Wei-Chih Hung, Chun-Yen Chu, Yi-Leh Wu, and Cheng-Yuan Tang, "Map/Reduce Affinity Propagation Clustering Algorithm," International Journal of Electronics and Electrical Engineering, Vol. 3, No. 4, pp. 311-317, August 2015. doi: 10.12720/ijeee.3.4.311-317
Array