How spam features change in twitter and the impact to machine learning based detection

Tingmin Wu; Derek Wang; Sheng Wen; Yang Xiang

doi:10.1007/978-3-319-72359-4_57

How spam features change in twitter and the impact to machine learning based detection

Tingmin Wu, Derek Wang, Sheng Wen, Yang Xiang

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

3 Citations (Scopus)

Abstract

Twitter Spam is a critical problem and current solution is mainly about machine learning based detection. However, recent studies found that the spam features are continuously changing day by day (called ‘Spam Drift’ problem), which may significantly affect the performance of the detection. In this paper, we carried out a real-data driven study to explored the ‘Spam Drift’ problem and its impact to machine learning based detection. Our study found that only a small group of spam features will continuously change. The results also suggested a counter-intuitive conclusion that the ‘Spam Drift’ problem does not have serious impact on spam detection Precision (SP) and non-spam detection Recall (NR), two metrics that industries prioritise in practice.

Original language	English
Title of host publication	Information Security Practice and Experience
Subtitle of host publication	13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings
Editors	Joseph K. Liu, Pierangela Samarati
Place of Publication	Cham Switzerland
Publisher	Springer
Pages	898-904
Number of pages	7
ISBN (Electronic)	9783319723594
ISBN (Print)	9783319723587
DOIs	https://doi.org/10.1007/978-3-319-72359-4_57
Publication status	Published - 2017
Externally published	Yes
Event	Information Security Practice and Experience Conference 2017 - Deakin Downtown, Melbourne, Australia Duration: 13 Dec 2017 → 15 Dec 2017 Conference number: 13th http://nsclab.org/ispec2017/ (Website) https://link.springer.com/book/10.1007/978-3-319-72359-4 (Proceedings)

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	10701
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	Information Security Practice and Experience Conference 2017
Abbreviated title	ISPEC 2017
Country/Territory	Australia
City	Melbourne
Period	13/12/17 → 15/12/17
Internet address	http://nsclab.org/ispec2017/ (Website) https://link.springer.com/book/10.1007/978-3-319-72359-4 (Proceedings)

Keywords

Security
Spam
Twitter

Access to Document

10.1007/978-3-319-72359-4_57

Cite this

Wu, T., Wang, D., Wen, S., & Xiang, Y. (2017). How spam features change in twitter and the impact to machine learning based detection. In J. K. Liu, & P. Samarati (Eds.), Information Security Practice and Experience: 13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings (pp. 898-904). (Lecture Notes in Computer Science ; Vol. 10701 ). Springer. https://doi.org/10.1007/978-3-319-72359-4_57

Wu, Tingmin ; Wang, Derek ; Wen, Sheng et al. / How spam features change in twitter and the impact to machine learning based detection. Information Security Practice and Experience: 13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings. editor / Joseph K. Liu ; Pierangela Samarati. Cham Switzerland : Springer, 2017. pp. 898-904 (Lecture Notes in Computer Science ).

@inproceedings{fd323eb11aae4b198908cd35de1eda24,

title = "How spam features change in twitter and the impact to machine learning based detection",

abstract = "Twitter Spam is a critical problem and current solution is mainly about machine learning based detection. However, recent studies found that the spam features are continuously changing day by day (called {\textquoteleft}Spam Drift{\textquoteright} problem), which may significantly affect the performance of the detection. In this paper, we carried out a real-data driven study to explored the {\textquoteleft}Spam Drift{\textquoteright} problem and its impact to machine learning based detection. Our study found that only a small group of spam features will continuously change. The results also suggested a counter-intuitive conclusion that the {\textquoteleft}Spam Drift{\textquoteright} problem does not have serious impact on spam detection Precision (SP) and non-spam detection Recall (NR), two metrics that industries prioritise in practice.",

keywords = "Security, Spam, Twitter",

author = "Tingmin Wu and Derek Wang and Sheng Wen and Yang Xiang",

year = "2017",

doi = "10.1007/978-3-319-72359-4_57",

language = "English",

isbn = "9783319723587",

series = "Lecture Notes in Computer Science ",

publisher = "Springer",

pages = "898--904",

editor = "Liu, {Joseph K.} and Pierangela Samarati",

booktitle = "Information Security Practice and Experience",

note = "Information Security Practice and Experience Conference 2017, ISPEC 2017 ; Conference date: 13-12-2017 Through 15-12-2017",

url = "http://nsclab.org/ispec2017/, https://link.springer.com/book/10.1007/978-3-319-72359-4",

}

Wu, T, Wang, D, Wen, S & Xiang, Y 2017, How spam features change in twitter and the impact to machine learning based detection. in JK Liu & P Samarati (eds), Information Security Practice and Experience: 13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings. Lecture Notes in Computer Science , vol. 10701 , Springer, Cham Switzerland, pp. 898-904, Information Security Practice and Experience Conference 2017, Melbourne, Victoria, Australia, 13/12/17. https://doi.org/10.1007/978-3-319-72359-4_57

How spam features change in twitter and the impact to machine learning based detection. / Wu, Tingmin; Wang, Derek; Wen, Sheng et al.
Information Security Practice and Experience: 13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings. ed. / Joseph K. Liu; Pierangela Samarati. Cham Switzerland: Springer, 2017. p. 898-904 (Lecture Notes in Computer Science ; Vol. 10701 ).

Research output: Chapter in Book/Report/Conference proceeding › Conference Paper › Research › peer-review

TY - GEN

T1 - How spam features change in twitter and the impact to machine learning based detection

AU - Wu, Tingmin

AU - Wang, Derek

AU - Wen, Sheng

AU - Xiang, Yang

N1 - Conference code: 13th

PY - 2017

Y1 - 2017

N2 - Twitter Spam is a critical problem and current solution is mainly about machine learning based detection. However, recent studies found that the spam features are continuously changing day by day (called ‘Spam Drift’ problem), which may significantly affect the performance of the detection. In this paper, we carried out a real-data driven study to explored the ‘Spam Drift’ problem and its impact to machine learning based detection. Our study found that only a small group of spam features will continuously change. The results also suggested a counter-intuitive conclusion that the ‘Spam Drift’ problem does not have serious impact on spam detection Precision (SP) and non-spam detection Recall (NR), two metrics that industries prioritise in practice.

AB - Twitter Spam is a critical problem and current solution is mainly about machine learning based detection. However, recent studies found that the spam features are continuously changing day by day (called ‘Spam Drift’ problem), which may significantly affect the performance of the detection. In this paper, we carried out a real-data driven study to explored the ‘Spam Drift’ problem and its impact to machine learning based detection. Our study found that only a small group of spam features will continuously change. The results also suggested a counter-intuitive conclusion that the ‘Spam Drift’ problem does not have serious impact on spam detection Precision (SP) and non-spam detection Recall (NR), two metrics that industries prioritise in practice.

KW - Security

KW - Spam

KW - Twitter

UR - http://www.scopus.com/inward/record.url?scp=85038083067&partnerID=8YFLogxK

U2 - 10.1007/978-3-319-72359-4_57

DO - 10.1007/978-3-319-72359-4_57

M3 - Conference Paper

AN - SCOPUS:85038083067

SN - 9783319723587

T3 - Lecture Notes in Computer Science

SP - 898

EP - 904

BT - Information Security Practice and Experience

A2 - Liu, Joseph K.

A2 - Samarati, Pierangela

PB - Springer

CY - Cham Switzerland

T2 - Information Security Practice and Experience Conference 2017

Y2 - 13 December 2017 through 15 December 2017

ER -

Wu T, Wang D, Wen S, Xiang Y. How spam features change in twitter and the impact to machine learning based detection. In Liu JK, Samarati P, editors, Information Security Practice and Experience: 13th International Conference, ISPEC 2017 Melbourne, VIC, Australia, December 13–15, 2017 Proceedings. Cham Switzerland: Springer. 2017. p. 898-904. (Lecture Notes in Computer Science ). doi: 10.1007/978-3-319-72359-4_57

How spam features change in twitter and the impact to machine learning based detection

Abstract

Publication series

Conference

Keywords

Access to Document

Other files and links

Cite this