Distribution-aware compressed full-text indexes

1Citations
Citations of this article
2Readers
Mendeley users who have this article in their library.
Get full text

Abstract

In this paper we address the problem of building a compressed self-index that, given a distribution for the pattern queries and a bound on the space occupancy, minimizes the expected query-time within that index-space bound. We solve this problem by exploiting a reduction to the problem of finding a minimum weight K-link path in a particular Directed Acyclic Graph. Interestingly enough, our solution is independent of the underlying compressed index in use. Our experiments compare this optimal strategy with several other standard approaches, showing its effectiveness in practice. © 2011 Springer-Verlag Berlin Heidelberg.

Cite

CITATION STYLE

APA

Ferragina, P., Sirén, J., & Venturini, R. (2011). Distribution-aware compressed full-text indexes. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 6942 LNCS, pp. 760–771). Springer Verlag. https://doi.org/10.1007/978-3-642-23719-5_64

Register to see more suggestions

Mendeley helps you to discover research relevant for your work.

Already have an account?

Save time finding and organizing research with Mendeley

Sign up for free