{"title":"Heterogeneous Attribute Reduction in Noisy System based on a Generalized Neighborhood Rough Sets Model","authors":"Siyuan Jing, Kun She","volume":51,"journal":"International Journal of Mathematical and Computational Sciences","pagesStart":351,"pagesEnd":357,"ISSN":"1307-6892","URL":"https:\/\/publications.waset.org\/pdf\/7597","abstract":"Neighborhood Rough Sets (NRS) has been proven to\r\nbe an efficient tool for heterogeneous attribute reduction. However,\r\nmost of researches are focused on dealing with complete and noiseless\r\ndata. Factually, most of the information systems are noisy, namely,\r\nfilled with incomplete data and inconsistent data. In this paper, we\r\nintroduce a generalized neighborhood rough sets model, called\r\nVPTNRS, to deal with the problem of heterogeneous attribute\r\nreduction in noisy system. We generalize classical NRS model with\r\ntolerance neighborhood relation and the probabilistic theory.\r\nFurthermore, we use the neighborhood dependency to evaluate the\r\nsignificance of a subset of heterogeneous attributes and construct a\r\nforward greedy algorithm for attribute reduction based on it.\r\nExperimental results show that the model is efficient to deal with noisy\r\ndata.","references":"[1] Pawlak Z., \"Rough sets.\", Theoretical Aspects of Reasoning about Data,\r\nKluwer, 1991.\r\n[2] G. Y. Wang, \"Rough Set Theory and Knowledge Discovery.\",\r\nXi-an:Xi-an Jiaotong University Press, 2001.\r\n[3] C. Cornelis, M. De Cock, A. Radzikowska, \"Vaguely Quantified Rough\r\nSets,\", Proc. 11th Int. Conf. on Rough Sets, Fuzzy Sets, Data Mining and\r\nGranular Computing (RSFDGrC2007), Lecture Notes in Artificial\r\nIntelligence 4482, 2007, pp: 87-94.\r\n[4] T. Y. Lin, Q. Liu, K J Huang, \"Rough sets neighborhood systems and\r\napproximation.\", In 15th International Symposium on Methodologies of\r\nIntelligent Systems, 1990.\r\n[5] Y. Y. Yao, \"Relational interpretation of neighborhood operators and\r\nrough set approximation operators.\", Information sciences, vol. 111,\r\nno.198, pp: 239-259, 1998.\r\n[6] W. Z. Wu, W. X. Zhang, \"Neighborhood operator systems and\r\napproximations.\", Information sciences, vol. 144, no.14, pp: 201-217,\r\n2002.\r\n[7] W. Ziarko, \"Set approximation quality measures in the variable precision\r\nrough set model.\", Soft Computing Systems, Management and\r\nApplications, pp: 442-452, 2001.\r\n[8] Q. H. Hu, D. R. Yu, Z. X. Xie, \"Numerical attribute reduction based on\r\nneighborhood granulation and rough approximation.\", Chinese Journal\r\nof software, vol. 19, no.3, pp.640\u2212649, 2008.\r\n[9] M. R. Alicja, L. Rolka, \"Variable Precision Fuzzy Rough Sets\",\r\nTransaction on Rough Sets, LNCS, 144-160, 2004.\r\n[10] D. J. Newman, S. Hettich, C. L. Blake, C. J. Merz, \"UCI Repository of\r\nMachine Learning Databases.\", University of California, Department of\r\nInformation and Computer Science, Irvine, CA, 1998.\r\n.\r\n[11] W. Ziarko, \" Variable precision rough set model \" , Journal of\r\nComputer and System Sciences, vol. 46, pp: 39-59, 1993.\r\n[12] U. Fayyad, K. Irani, \"Discrediting continuous attributes while learning\r\nBayesian networks.\", in 13th International Conference on Machine\r\nLearning, Morgan Kaufmann, 1996, pp: 157- 165.","publisher":"World Academy of Science, Engineering and Technology","index":"Open Science Index 51, 2011"}