Skip to main content

Sampling Techniques for Statistical Databases

  • Living reference work entry
  • First Online:
Encyclopedia of Database Systems
  • 117 Accesses

Definition

A sampling technique is a method by which one inspects only a small portion of data from a database to reduce the time to compute an aggregate query, but simultaneously ensuring that result computed on the sample faithfully represents the true results of the query for the entire data population.

Example

Acceptance-Rejection sampling(AR sampling) is sampling technique.

Key Points

Sampling is used in a database for different reasons such as (i) to estimate the results of aggregate queries (e.g., SUM, COUNT, orAVERAGE), (ii) to retrieve a sample of records from a database query for subsequent processing, (iii) for internal use by the query optimizer for selectivity estimation, (iv)to provide privacy protection for records on individuals contained in statistical databases. It has been determined that fixed size random sampling of data does not yield a true representation of the population. Acceptance/rejection (A/R) samplingis used to construct weighted samples in which the...

This is a preview of subscription content, log in via an institution to check access.

Access this chapter

Institutional subscriptions

Recommended Reading

  1. Olken F, Rotem D. Random sampling from databases: a survey. Stat Comput. 1995;5:25–42.

    Article  Google Scholar 

Download references

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Amarnath Gupta .

Editor information

Editors and Affiliations

Section Editor information

Rights and permissions

Reprints and permissions

Copyright information

© 2016 Springer Science+Business Media LLC

About this entry

Cite this entry

Gupta, A. (2016). Sampling Techniques for Statistical Databases. In: Liu, L., Özsu, M. (eds) Encyclopedia of Database Systems. Springer, New York, NY. https://doi.org/10.1007/978-1-4899-7993-3_1293-2

Download citation

  • DOI: https://doi.org/10.1007/978-1-4899-7993-3_1293-2

  • Received:

  • Accepted:

  • Published:

  • Publisher Name: Springer, New York, NY

  • Online ISBN: 978-1-4899-7993-3

  • eBook Packages: Springer Reference Computer SciencesReference Module Computer Science and Engineering

Publish with us

Policies and ethics