High-Order Entropy Compressed Bit Vectors with Rank/Select

Author: Beskers Kai   Fischer Johannes  

Publisher: MDPI

E-ISSN: 1999-4893|7|4|608-620

ISSN: 1999-4893

Source: Algorithms, Vol.7, Iss.4, 2014-11, pp. : 608-620

Access to resources Favorite

Disclaimer: Any content in publications that violate the sovereignty, the constitution or regulations of the PRC is not accepted or approved by CNPIEC.

Previous Menu Next

Abstract

We design practical implementations of data structures for compressing bit-vectors to support efficient rank-queries (counting the number of ones up to a given point). Unlike previous approaches, which either store the bit vectors plainly, or focus on compressing bit-vectors with low densities of ones or zeros, we aim at low entropies of higher order, for example 101010 10 . Our implementations achieve very good compression ratios, while showing only a modest increase in query time.