2010 | OriginalPaper | Buchkapitel
Using Related Queries to Improve Web Search Results Ranking
verfasst von : Georges Dupret, Ricardo Zilleruelo-Ramos, Sumio Fujita
Erschienen in: String Processing and Information Retrieval
Verlag: Springer Berlin Heidelberg
Aktivieren Sie unsere intelligente Suche, um passende Fachinhalte oder Patente zu finden.
Wählen Sie Textabschnitte aus um mit Künstlicher Intelligenz passenden Patente zu finden. powered by
Markieren Sie Textabschnitte, um KI-gestützt weitere passende Inhalte zu finden. powered by
There are numerous queries for which search engine results are not satisfactory. For instance, the user may submit an ambiguous or miss-spelled query; or there might be a mismatch between query and document vocabulary, or even character set in some languages. Different automatic methods for query rewriting / refinement have been proposed in the literature, but little work has been done on how to combine the results of these rewrites to find relevant documents. In this paper, we review some techniques efficient enough to be computed online and we discuss their respective assumptions. We also propose and discuss a new model that is theoretically more appealing while still computationally very efficient. Our experiments show that all methods manage to improve the ranking of a leading commercial search engine.