Affinity Propagation with Background Knowledge using Pairwise Constraints

International Journal of Computer Science and Engineering
© 2017 by SSRG - IJCSE Journal
Volume 4 Issue 2
Year of Publication : 2017
Authors : Saravanakumar.R, Dr.C.Nandini

How to Cite?

Saravanakumar.R, Dr.C.Nandini, "Affinity Propagation with Background Knowledge using Pairwise Constraints," SSRG International Journal of Computer Science and Engineering , vol. 4,  no. 2, pp. 1-3 , 2017. Crossref,


Data mining is the process of identifying and extracting hidden patterns and information from large databases and warehouses. Incorporating pairwise constraints into clustering algorithms is an emerging research area for machine learning and data mining communities. Already various algorithms exist to combine relative similarities between clusters from different viewpoints. But they suffer from duplicates in clusters and also lesser relevancy. The proposed Affinity propagation clustering algorithm uses semi-supervised learning to avoid data redundancy from input strings and ensures quicker retrieval. Final Clusters contain unique and relevant data. Semi-supervised learning falls between unsupervised learning (without any label training data) and supervised learning (with completely labelled training data). Thus the hybrid algorithms provides performance enhancement over its existing counterparts. Further large amount of input data can be processed precisely and even various alternative forms of similar output data can be retrieved. Hence the highest degree of accuracy can be achieved in clustering data and retrieval of the same by the improved affinity propagation algorithm.


 Affinity propagation, clusters, pairwise constraints, semi-supervised learning.


[1] Antonio Augusto Chaves, Luiz Antonio Nogueira Lorena, „Clustering Search Algorithm for the Capacitated Centered Clustering Problem.
[2] Basu.S, Bilenko.M, and Mooney.RJ, (2004) „A Probabilistic Framework for Semi-Supervised Clustering, Proc. ACM SIGKDD Intl Conf. Knowledge Discovery and Data Mining, pp. 59-68.
[3] Basu.S, Banerjee.A, and Mooney.R.J, (2004) „Acive Semi- Supervision for Pairwise Constraints, ICDM 04.
[4] Bilenko.M, Basu.S, and MooneyR.J, (2004) „Integrating Constraints and Metric Learning in Semi-Supervised Clustering, Proc. Intl Conf. Machine Learning, pp. 81-88.
[5] Dhillon.I.S, Mallela.S, and ModhaD.S, (2003) „Information- Theoretic Co-Clustering, Proc. Ninth ACM SIGKDD Intl Conf. Knowledge Discovery and Data Mining (KDD), pp. 89- 98.
[6] Ding.C, He.X, Zha.H, Gu.M, and Simon.H, (2001) „A Min- Max Cut Algorithm for Graph Partitioning and Data Clustering, Proc. IEEE Intl Conf. Data Mining (ICDM), pp. 107-114.
[7] Ester.M, Ge.R, Gao.B.J, Hu.Z, and Ben-Moshe.B, (2006) „Joint Cluster Analysis of Attribute Data and Relationship Data: The Connected K-Center Problem, Proc. SIAM Int‟l Conf. Data Mining, pp. 25-46.
[8] Hoi.S.C.H, Liu.W, Lyu.M.R, and Ma.W.Y, (2006) „Learning Distance Metrics with Contextual Constraints for Image Retrieval, Proc. IEEE CS Conf. Computer Vision and Pattern Recognition, pp. 2072-2078.
[9] Kulis.B, Basu.S, Dhillon.I, and Mooney.R, (2005) “Semi- Supervised Graph Clustering: A Kernel Approach,” Proc. Int‟l Conf. Machine Learning, pp. 457-464.
[10] Law.M, Topchy.A, and Jain.A.K, (2005) „Model-Based Clustering with Probabilistic Constraints, Proc. SIAM Int‟l Conf. Data Mining, pp. 641-645.
[11] Wagstaff.k, Cardie.C, and Schroedl.S, (2001) „Constrained KMeans Clustering with Background Knowledge, Proc. Int‟l Conf. Machine Learning, pp. 577-584.
[12] Xing.E.P, Ng. A.Y,Jordan. M.I, and Russell.S, (2003) „Distance Metric Learning with Application to Clustering with Side-Information, Advances in Neural Information Processing Systems, vol. 15, pp. 521-528.
[13] Zha.H, He.X, Ding.C, Simon.H, and Gu.M, (2001) „Spectral Relaxation for K-Means Clustering, Proc. Neural Info. Processing Systems (NIPS), pp. 1057-1064.