An Efficient Search Algorithm for High-Dimensional Indexing Using Cell Based MBR

  • Bo-Hyun Wang
  • Byung-Wook Lee
Conference paper
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3980)


Among the many issues in high dimensional index structures using Minimum Bounding Rectangle(MBR), the reduction of fan-out and increase of overlapping area are the key factors in reduction of search speed. It is known that the usage of only minimum and maximum distance in MBR’s pruning process lowers the accuracy of search. In this paper, we present an index structure using cell based MBR in which fan-out gets increased and overlapping is avoided, and a search algorithm which reflects the distribution status of data in MBR to the search. The proposed index structure produces MBR as Vector Approximation-file(VA-file)’s cell units and produces child-MBR by dividing cells. The search algorithm raises the search accuracy by executing pruning using centroid of values included in MBR other than the minimum and maximum distance of cell based MBR and query vector in the k-nn query concerned. Through experiment, we find that the proposed search algorithm has improved its search speed and its accuracy in comparison with existing algorithm.


Feature Vector Search Algorithm Leaf Node Index Structure Data Node 


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Sakurai., Y., Yoshikawa., M., Uemura., S., Kojima, H.: The A-tree: An Index Structure for High Dimensional Spaces Using Relative Approximation. In: Proceedings on the 26th VLDB Conference, Egypt, pp. 516–526 (2000)Google Scholar
  2. 2.
    Sakurai., Y., Yoshikawa., M., Uemura., S., Kojima, H.: The A-tree: An Index Structure for High-Dimensional Spaces Using Relative Approximation. Technical report, Nara Institute of Science and Technology, pp.1–22 (2000)Google Scholar
  3. 3.
    Beckmann, N., Kriegel, H.-P., Schneider, R., Seeger, B.: The R*-tree:An Efficient and Robust Access Method for Points and Rectangles. In: Proceedings on ACM SIGMOD International Conference on Management of Data, Atlantic City, NJ, pp. 322–331 (1990)Google Scholar
  4. 4.
    Katayama., N., Satoh, S.: The SR-tree: An index structure for high-dimensional nearest neighbor queries. In: Proceedings on the ACM SIGMOD International Conference on Management of Data, Tucson, Arizon USA, vol. 26(2), pp. 369–380 (1997)Google Scholar
  5. 5.
    Weber, R., Schek, H., Blott, S.: A quantitative analysis and performance study for similarity-search methods in highdimensional spaces. In: Proceedings on the 24th VLDB Conference, New York City, pp. 194–205 (1998)Google Scholar
  6. 6.
    White., D.A., Jain, R.: Similarity Indexing with the SS-tree. In: Proceedings Of the 12th International Conference on Data Engineering, New Orleans, USA, pp. 516–523 (1996)Google Scholar
  7. 7.
    Roussopoulos., N., Kelley., S., Vincent, F.: Nearest Neighbor Queries. In: Proceedings ACM SIGMOD, San Jose, USA, pp. 71–79 (1995)Google Scholar
  8. 8.
    Berchtold, S., Keim, D.A., Kriegel, H.-P.: The X-tree: An Index Structure for High-Dimensional Data. In: Proceedings on the 22nd VLDB Conference, Bombay, pp. 28–39 (1996)Google Scholar
  9. 9.
    Lin, K., Jagadish, H.V., Faloutsos, C.: The TV-tree: An Index Structure for High-Dimensional Data. VLDB Journal 3(4), 517–542 (1994)CrossRefGoogle Scholar
  10. 10.
    Berchtold, S., Bohm, C., Kriegel, H.-P.: The Pyramid-Technique: Towards Breaking the Curse of Dimensionality. In: ACM SIGMOD, Seattle, vol. 27(2), pp. 142–153 (1998)Google Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Bo-Hyun Wang
    • 1
  • Byung-Wook Lee
    • 1
  1. 1.College of SoftwareKyungwon UniversityGyeonggi-doSouth Korea

Personalised recommendations