Skip to main navigation Skip to search Skip to main content

Inducing codes from examples

  • Stony Brook University

Research output: Chapter in Book/Report/Conference proceedingConference contributionpeer-review

Abstract

We propose a data compression algorithm which automatically analyzes a collection of examples to identify the set of strings which would be most useful to encode the examples. There is considerable subtlety in identifying the most useful strings, since the problem is NP-complete, but we have developed analysis and encoding/decoding heuristics which construct excellent codes. In this paper, we describe our algorithm and experimental results on four different special domains: mailing addresses, Fortran programs, weather radar images, and UNIX manual pages. In each of these domains, our method significantly outperformed such standard compression algorithms as Huffman codes and LZW.

Original languageEnglish
Title of host publicationData Compression Conference 1991
PublisherInstitute of Electrical and Electronics Engineers Inc.
Pages267-276
Number of pages10
ISBN (Electronic)0818692022
DOIs
StatePublished - 1991
Event1991 Data Compression Conference, DCC 1991 - Snowbird, United States
Duration: Apr 8 1991Apr 11 1991

Publication series

NameData Compression Conference Proceedings
Volume1991-April
ISSN (Print)1068-0314

Conference

Conference1991 Data Compression Conference, DCC 1991
Country/TerritoryUnited States
CitySnowbird
Period04/8/9104/11/91

Fingerprint

Dive into the research topics of 'Inducing codes from examples'. Together they form a unique fingerprint.

Cite this