Context-Aware Systems and Applications, and Nature of Computation and Communication. 6th International Conference, ICCASA 2017, and 3rd International Conference, ICTCC 2017, Tam Ky, Vietnam, November 23-24, 2017, Proceedings

Research Article

Fragmentation in Distributed Database Design Based on KR Rough Clustering Technique

Download
175 downloads
  • @INPROCEEDINGS{10.1007/978-3-319-77818-1_16,
        author={Van Luong and Van Le and Van Doan},
        title={Fragmentation in Distributed Database Design Based on KR Rough Clustering Technique},
        proceedings={Context-Aware Systems and Applications, and Nature of Computation and Communication. 6th International Conference, ICCASA 2017, and 3rd International Conference, ICTCC 2017, Tam Ky, Vietnam, November 23-24, 2017, Proceedings},
        proceedings_a={ICCASA \& ICTCC},
        year={2018},
        month={3},
        keywords={Rough set theory Vertical fragmentation Rough cluster Cluster focus},
        doi={10.1007/978-3-319-77818-1_16}
    }
    
  • Van Luong
    Van Le
    Van Doan
    Year: 2018
    Fragmentation in Distributed Database Design Based on KR Rough Clustering Technique
    ICCASA & ICTCC
    Springer
    DOI: 10.1007/978-3-319-77818-1_16
Van Luong1,*, Van Le2,*, Van Doan3,*
  • 1: Pham Van Dong University
  • 2: Da Nang University of Education
  • 3: Vietnamese Academy of Science and Technology
*Contact email: nghia.itq@gmail.com, levansupham2004@yahoo.com, dvban@ioit.ac.vn

Abstract

Knowledge mining according to rough set approach is an effective method for large datasets containing many different types of data. Rough clustering, as in rough set theory, using lower approximation and upper approximation, allows objects to belong to multiple clusters in a dataset. KR Rough Clustering Technique (K-Means Rough) we propose in this paper follows k-Means primitive clustering algorithm improvement approach by combining distance, similarity with upper approximation and lower approximation. In particular, appropriate focuses will be calculated to determine whether an object will be assigned to lower approximation or upper approximation of each cluster.