English

Grammar compression with probabilistic context-free grammar

Data Structures and Algorithms 2020-03-19 v1

Abstract

We propose a new approach for universal lossless text compression, based on grammar compression. In the literature, a target string TT has been compressed as a context-free grammar GG in Chomsky normal form satisfying L(G)={T}L(G) = \{T\}. Such a grammar is often called a \emph{straight-line program} (SLP). In this paper, we consider a probabilistic grammar GG that generates TT, but not necessarily as a unique element of L(G)L(G). In order to recover the original text TT unambiguously, we keep both the grammar GG and the derivation tree of TT from the start symbol in GG, in compressed form. We show some simple evidence that our proposal is indeed more efficient than SLPs for certain texts, both from theoretical and practical points of view.

Keywords

Cite

@article{arxiv.2003.08097,
  title  = {Grammar compression with probabilistic context-free grammar},
  author = {Hiroaki Naganuma and Diptarama Hendrian and Ryo Yoshinaka and Ayumi Shinohara and Naoki Kobayashi},
  journal= {arXiv preprint arXiv:2003.08097},
  year   = {2020}
}

Comments

11 pages, 3 figures, accepted for poster presentation at DCC 2020

R2 v1 2026-06-23T14:18:22.315Z