Top-k Algorithm Based on Extraction

Conference paper
Part of the Advances in Intelligent and Soft Computing book series (AINSC, volume 144)

Abstract

Algorithms for top-k query are widely used in massive data query, which return k most important objects based on aggregate functions. The classical Threshold Algorithm (TA) is one of the most famous algorithms for top-k query. It requires sequential and random accesses to the lists. The time cost of TA will be very high when data is massive. This paper proposes a new algorithm TABE (Top-k Algorithm Based on Extraction) to minimize the query time. TABE first extracts the objects which have higher ranking on each attribute, and then execute the Threshold Algorithm on these objects. Test results show that TABE has high accuracy to meet the general query requirements, and the experimental results of comparing TABE with NRA (No Random Accesses) show that our proposed algorithm TABE can largely reduce the query time.

Preview

Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.

References

  1. 1.
    Fagin, R., Lotem, A., Naor, M.: Optimal aggregation algorithms for middleware. In: Proceedings of the 20th ACM SIGACT-SIGMOD-SIGART Symposium on Principles of Database Systems (PODS 2001), California, USA, pp. 102–113 (2001)Google Scholar
  2. 2.
    Fagin, R., Lotem, A., Naor, M.: Optimal aggregation algorithms for middleware. Journal of Computer and System Sciences 66(4), 614–656 (2003)MathSciNetMATHCrossRefGoogle Scholar
  3. 3.
    Pang, H., Ding, X., Zheng, B.: Efficient processing of exact Top-k queries over disk-resident sorted lists. VLDB Journal 19(3), 437–456 (2010)CrossRefGoogle Scholar
  4. 4.
    Han, X., Yang, D., Li, J.: TKEP: an efficient Top-k query processing algorithm on massive data. Chinese Journal of Computers 33(8), 1405–1417 (2010)MathSciNetCrossRefGoogle Scholar
  5. 5.
    Mamoulis, N., Cheng, K.H., Yiu, M.L., Cheung, D.W.: Efficient aggregation of ranked inputs. In: Proceedings of the 22nd International Conference on Data Engineering (ICDE 2006), Atlanta, GA, USA, pp. 72–83 (2006)Google Scholar

Copyright information

© Springer-Verlag GmbH Berlin Heidelberg 2012

Authors and Affiliations

  1. 1.College of Computer ScienceNanjing University of Posts and TelecommunicationsNanjingChina
  2. 2.Department of Information Engineering and Computer SceinceUniversity of TrentoTrentoItaly

Personalised recommendations