2016 International Conference on Big Data and Smart Computing (BigComp) (2016)
Hong Kong, China
Jan. 18, 2016 to Jan. 20, 2016
Jong Myoung Kim , School of Computing, KAIST, Korea
Hancheol Park , Department of Knowledge Service Engineering, KAIST, Korea
Gahgene Gweon , Department of Knowledge Service Engineering, KAIST, Korea
Jeong Hur , Knowledge Mining Research Team, ETRI, Korea
We apply the notion of "popularity" in machine-generated sentence evaluation to the queries used to search for documents. Our intuition is that queries composed of popular terms obtain more relevant documents and increase the probability that these documents contain the desired results. We measure the popularity of a query by analyzing a massive online document repository, Korean Wikipedia. To verify the influence of query popularity on search results, we conduct experiments to measure the mean reciprocal rank and precision-at-k, and perform an individual comparison of search term pairs. Through these experiments, we demonstrate that better search results can be obtained by considering the popularity of the query.
Context, Encyclopedias, Electronic publishing, Internet, Search engines, Mathematical model
J. M. Kim, H. Park, G. Gweon and J. Hur, "The correlation between search quality and query popularity," 2016 International Conference on Big Data and Smart Computing (BigComp)(BIGCOMP), Hong Kong, China, 2016, pp. 353-356.