Content-based mobile spam classification using stylistically motivated features

Dae Neung Sohn, Jung Tae Lee, Kyoung Soo Han, Hae-Chang Rim

Research output: Contribution to journalArticle

21 Citations (Scopus)

Abstract

The feature of brevity in mobile phone messages makes it difficult to distinguish lexical patterns to identify spam. This paper proposes a novel approach to spam classification of extremely short messages using not only lexical features that reflect the content of a message but new stylistic features that indicate the manner in which the message is written. Experiments on two mobile phone message collections in two different languages show that the approach outperforms previous content-based approaches significantly, regardless of language.

Original languageEnglish
Pages (from-to)364-369
Number of pages6
JournalPattern Recognition Letters
Volume33
Issue number3
DOIs
Publication statusPublished - 2012 Feb 1

Fingerprint

Mobile phones
Experiments

Keywords

  • Mobile spam classification
  • Stylistic features
  • Text messaging service

ASJC Scopus subject areas

  • Software
  • Artificial Intelligence
  • Computer Vision and Pattern Recognition
  • Signal Processing

Cite this

Content-based mobile spam classification using stylistically motivated features. / Sohn, Dae Neung; Lee, Jung Tae; Han, Kyoung Soo; Rim, Hae-Chang.

In: Pattern Recognition Letters, Vol. 33, No. 3, 01.02.2012, p. 364-369.

Research output: Contribution to journalArticle

Sohn, Dae Neung ; Lee, Jung Tae ; Han, Kyoung Soo ; Rim, Hae-Chang. / Content-based mobile spam classification using stylistically motivated features. In: Pattern Recognition Letters. 2012 ; Vol. 33, No. 3. pp. 364-369.
@article{e9b98a2f76b445b28b4fa33f80d8c9dd,
title = "Content-based mobile spam classification using stylistically motivated features",
abstract = "The feature of brevity in mobile phone messages makes it difficult to distinguish lexical patterns to identify spam. This paper proposes a novel approach to spam classification of extremely short messages using not only lexical features that reflect the content of a message but new stylistic features that indicate the manner in which the message is written. Experiments on two mobile phone message collections in two different languages show that the approach outperforms previous content-based approaches significantly, regardless of language.",
keywords = "Mobile spam classification, Stylistic features, Text messaging service",
author = "Sohn, {Dae Neung} and Lee, {Jung Tae} and Han, {Kyoung Soo} and Hae-Chang Rim",
year = "2012",
month = "2",
day = "1",
doi = "10.1016/j.patrec.2011.10.017",
language = "English",
volume = "33",
pages = "364--369",
journal = "Pattern Recognition Letters",
issn = "0167-8655",
publisher = "Elsevier",
number = "3",

}

TY - JOUR

T1 - Content-based mobile spam classification using stylistically motivated features

AU - Sohn, Dae Neung

AU - Lee, Jung Tae

AU - Han, Kyoung Soo

AU - Rim, Hae-Chang

PY - 2012/2/1

Y1 - 2012/2/1

N2 - The feature of brevity in mobile phone messages makes it difficult to distinguish lexical patterns to identify spam. This paper proposes a novel approach to spam classification of extremely short messages using not only lexical features that reflect the content of a message but new stylistic features that indicate the manner in which the message is written. Experiments on two mobile phone message collections in two different languages show that the approach outperforms previous content-based approaches significantly, regardless of language.

AB - The feature of brevity in mobile phone messages makes it difficult to distinguish lexical patterns to identify spam. This paper proposes a novel approach to spam classification of extremely short messages using not only lexical features that reflect the content of a message but new stylistic features that indicate the manner in which the message is written. Experiments on two mobile phone message collections in two different languages show that the approach outperforms previous content-based approaches significantly, regardless of language.

KW - Mobile spam classification

KW - Stylistic features

KW - Text messaging service

UR - http://www.scopus.com/inward/record.url?scp=84255178470&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84255178470&partnerID=8YFLogxK

U2 - 10.1016/j.patrec.2011.10.017

DO - 10.1016/j.patrec.2011.10.017

M3 - Article

AN - SCOPUS:84255178470

VL - 33

SP - 364

EP - 369

JO - Pattern Recognition Letters

JF - Pattern Recognition Letters

SN - 0167-8655

IS - 3

ER -