Distributed Processing and Development of New Retrieval Performance Measures for Query Reformulation Algorithms


The Transactions of the Korea Information Processing Society (1994 ~ 2000), Vol. 4, No. 4, pp. 963-972, Apr. 1997
10.3745/KIPSTE.1997.4.4.963,   PDF Download:

Abstract

In information retrieval, query reformulation algorithms construct queries from a set of initial input and feedback documents, and retrieval performance can be varied by different sets of input documents. In this study, we developed a criterion for measuring the performance sensitivity of query reformulation algorithms to input sets. In addition, we also propose a way of measuring the changes in retrieved area (CIRA) during query reformulation. We compared CIRAs of query reformulation algorithms (i.e., query tree, DNF method, and Dillon's method) using three test sets : the CACM. CISI, and Medlars. In the experiments, the query tree showed the highest decreasing CIRAs during reformulations, which means the fastest convergence rate to an output set. For sensitivity analysis, the query tree scored the highest sensitivity to different input sets even though its differences to the other algorithms are very small.


Statistics
Show / Hide Statistics

Statistics (Cumulative Counts from September 1st, 2017)
Multiple requests among the same browser session are counted as one view.
If you mouse over a chart, the values of data points will be shown.


Cite this article
[IEEE Style]
K. N. Ho, "Distributed Processing and Development of New Retrieval Performance Measures for Query Reformulation Algorithms," The Transactions of the Korea Information Processing Society (1994 ~ 2000), vol. 4, no. 4, pp. 963-972, 1997. DOI: 10.3745/KIPSTE.1997.4.4.963.

[ACM Style]
Kim Nam Ho. 1997. Distributed Processing and Development of New Retrieval Performance Measures for Query Reformulation Algorithms. The Transactions of the Korea Information Processing Society (1994 ~ 2000), 4, 4, (1997), 963-972. DOI: 10.3745/KIPSTE.1997.4.4.963.