Introduction of N-gram into a run-length encoding based ASCII art extraction method

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

As ASCII arts can be noise for natural language processing, ASCII art extraction methods can be used to remove them from text. A run-length encoding (RLE) based ASCII art extraction method proposed in our papers uses compression ratio by RLE for recognition of ASCII arts as ASCII arts tend to be compressed small by RLE and non-ASCII arts do not. It is because same characters tend to occur successively in ASCII arts but they do not in non-ASCII arts. Small ASCII arts, however, are not compressed as small as large ASCII arts. In this paper, we add the occurrence number of n-gram of ASCII arts in text into the RLE-based method as a new text attribute to cope with small ASCII arts. Our experimental results show that the new attribute improves the F-measure but it adds language-dependency into the RLE-based method though it is desirable that ASCII art extraction methods are language- independent.

Original languageEnglish
Title of host publicationLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
PublisherSpringer Verlag
Pages28-39
Number of pages12
Volume9396
ISBN (Print)9783319247991
DOIs
Publication statusPublished - 2015
Event15th International Conference on Current Trends in Web Engineering, ICWE 2015 Workshops NLPIT, PEWET, SoWEMine - Rotterdam, Netherlands
Duration: 2015 Jun 232015 Jun 26

Publication series

NameLecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)
Volume9396
ISSN (Print)03029743
ISSN (Electronic)16113349

Other

Other15th International Conference on Current Trends in Web Engineering, ICWE 2015 Workshops NLPIT, PEWET, SoWEMine
CountryNetherlands
CityRotterdam
Period15/6/2315/6/26

Fingerprint

Processing

Keywords

  • ASCII art
  • Natural language processing
  • Pattern recognition

ASJC Scopus subject areas

  • Computer Science(all)
  • Theoretical Computer Science

Cite this

Suzuki, T. (2015). Introduction of N-gram into a run-length encoding based ASCII art extraction method. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics) (Vol. 9396, pp. 28-39). (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9396). Springer Verlag. https://doi.org/10.1007/978-3-319-24800-4_3

Introduction of N-gram into a run-length encoding based ASCII art extraction method. / Suzuki, Tetsuya.

Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9396 Springer Verlag, 2015. p. 28-39 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics); Vol. 9396).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Suzuki, T 2015, Introduction of N-gram into a run-length encoding based ASCII art extraction method. in Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). vol. 9396, Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), vol. 9396, Springer Verlag, pp. 28-39, 15th International Conference on Current Trends in Web Engineering, ICWE 2015 Workshops NLPIT, PEWET, SoWEMine, Rotterdam, Netherlands, 15/6/23. https://doi.org/10.1007/978-3-319-24800-4_3
Suzuki T. Introduction of N-gram into a run-length encoding based ASCII art extraction method. In Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9396. Springer Verlag. 2015. p. 28-39. (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)). https://doi.org/10.1007/978-3-319-24800-4_3
Suzuki, Tetsuya. / Introduction of N-gram into a run-length encoding based ASCII art extraction method. Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics). Vol. 9396 Springer Verlag, 2015. pp. 28-39 (Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)).
@inproceedings{7e77c8c5bc244c1fa4a1c34dee440b00,
title = "Introduction of N-gram into a run-length encoding based ASCII art extraction method",
abstract = "As ASCII arts can be noise for natural language processing, ASCII art extraction methods can be used to remove them from text. A run-length encoding (RLE) based ASCII art extraction method proposed in our papers uses compression ratio by RLE for recognition of ASCII arts as ASCII arts tend to be compressed small by RLE and non-ASCII arts do not. It is because same characters tend to occur successively in ASCII arts but they do not in non-ASCII arts. Small ASCII arts, however, are not compressed as small as large ASCII arts. In this paper, we add the occurrence number of n-gram of ASCII arts in text into the RLE-based method as a new text attribute to cope with small ASCII arts. Our experimental results show that the new attribute improves the F-measure but it adds language-dependency into the RLE-based method though it is desirable that ASCII art extraction methods are language- independent.",
keywords = "ASCII art, Natural language processing, Pattern recognition",
author = "Tetsuya Suzuki",
year = "2015",
doi = "10.1007/978-3-319-24800-4_3",
language = "English",
isbn = "9783319247991",
volume = "9396",
series = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",
publisher = "Springer Verlag",
pages = "28--39",
booktitle = "Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)",

}

TY - GEN

T1 - Introduction of N-gram into a run-length encoding based ASCII art extraction method

AU - Suzuki, Tetsuya

PY - 2015

Y1 - 2015

N2 - As ASCII arts can be noise for natural language processing, ASCII art extraction methods can be used to remove them from text. A run-length encoding (RLE) based ASCII art extraction method proposed in our papers uses compression ratio by RLE for recognition of ASCII arts as ASCII arts tend to be compressed small by RLE and non-ASCII arts do not. It is because same characters tend to occur successively in ASCII arts but they do not in non-ASCII arts. Small ASCII arts, however, are not compressed as small as large ASCII arts. In this paper, we add the occurrence number of n-gram of ASCII arts in text into the RLE-based method as a new text attribute to cope with small ASCII arts. Our experimental results show that the new attribute improves the F-measure but it adds language-dependency into the RLE-based method though it is desirable that ASCII art extraction methods are language- independent.

AB - As ASCII arts can be noise for natural language processing, ASCII art extraction methods can be used to remove them from text. A run-length encoding (RLE) based ASCII art extraction method proposed in our papers uses compression ratio by RLE for recognition of ASCII arts as ASCII arts tend to be compressed small by RLE and non-ASCII arts do not. It is because same characters tend to occur successively in ASCII arts but they do not in non-ASCII arts. Small ASCII arts, however, are not compressed as small as large ASCII arts. In this paper, we add the occurrence number of n-gram of ASCII arts in text into the RLE-based method as a new text attribute to cope with small ASCII arts. Our experimental results show that the new attribute improves the F-measure but it adds language-dependency into the RLE-based method though it is desirable that ASCII art extraction methods are language- independent.

KW - ASCII art

KW - Natural language processing

KW - Pattern recognition

UR - http://www.scopus.com/inward/record.url?scp=84951019227&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84951019227&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-24800-4_3

DO - 10.1007/978-3-319-24800-4_3

M3 - Conference contribution

SN - 9783319247991

VL - 9396

T3 - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

SP - 28

EP - 39

BT - Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)

PB - Springer Verlag

ER -