A Method for Precision Improvement Based on Core Query Clusters and Term Proximity


The KIPS Transactions:PartB , Vol. 17, No. 5, pp. 399-404, Oct. 2010
10.3745/KIPSTB.2010.17.5.399,   PDF Download:

Abstract

In this paper, we propose a method for precision improvement based on core clusters and term proximity. The method is composed by three steps. The initial retrieval documents are clustered based on query term combination, which occurred in the document. Core clusters are selected by using proximity between query terms. Then, the documents in core clusters are reranked based on context information of query. On TREC AP test collection, experimental results in precision at the top documents(P@100) show that the proposed method improved 11.2% over the language model.


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. H. Jang and K. S. Lee, "A Method for Precision Improvement Based on Core Query Clusters and Term Proximity," The KIPS Transactions:PartB , vol. 17, no. 5, pp. 399-404, 2010. DOI: 10.3745/KIPSTB.2010.17.5.399.

[ACM Style]
Kye Hun Jang and Kyung Soon Lee. 2010. A Method for Precision Improvement Based on Core Query Clusters and Term Proximity. The KIPS Transactions:PartB , 17, 5, (2010), 399-404. DOI: 10.3745/KIPSTB.2010.17.5.399.