Classification of web robots: An empirical study based on over one billion requests

Junsup Lee; Sungdeok Cha; Dongkun Lee; Hyungkyu Lee

doi:10.1016/j.cose.2009.05.004

Classification of web robots: An empirical study based on over one billion requests

Junsup Lee, Sungdeok Cha, Dongkun Lee, Hyungkyu Lee

Department of Computer Science and Engineering

Research output: Contribution to journal › Article › peer-review

34 Citations (Scopus)

Abstract

Many studies on detection and classification of web robots have focused their attention mostly on text crawlers, and empirical experiments used relatively small data collected at universities. In this paper, we analyzed more than one billion requests to www.microsoft.com in 24 h. Web logs were made anonymous to eliminate potential privacy concerns while preserving essential characteristics (e.g., frequency, queries, etc). We have developed an effective characterization metrics, based on workload characteristics and resource types, in detecting and classifying various web robots including text crawlers, link checkers, and icon crawlers. As expected, web robot behavior was clearly different from that of typical interactive users, and different types of web robots also exhibited different characteristics. However, comparison of the similar type of web robots, text crawlers in particular, revealed different characteristics, thereby enabling characterization with reasonably high confidence level. We divided various feature metrics into five groups, and effectiveness of each group in classification is shown in polar diagram in the decreasing order of effectiveness in the clockwise direction. One can use the findings to classify likely identify of unknown web robots, and organizations can develop appropriate measures to deal with them. Our analysis is based on recent web log data collected at one of the best known site which offers truly global service. Crown

Original language	English
Pages (from-to)	795-802
Number of pages	8
Journal	Computers and Security
Volume	28
Issue number	8
DOIs	https://doi.org/10.1016/j.cose.2009.05.004
Publication status	Published - 2009 Nov

Bibliographical note

Funding Information:
Authors would like to thank Microsoft Corporation, MSRA UR in particular, for its generous support without which research reported in this paper could not have been performed. MSRA provided us raw data as well as research gift. Authors also acknowledge support provided by the following organizations: Defense Acquisition Program Administration and Agency for Defense Development; Korea SW Industry Promotion Agency through Software Engineering Technologies Development and Experts Education program; MKE (Ministry of Knowledge Economy) and IITA, Korea through ITRC (Information Technology Research Centre) support program IITA-2008-(C1090-0801-0032); MKE/MCST/IITA program 2008-S-025-02 at ETRI; and a research grant from Korea University.

Keywords

Web robot characterization
Web robot classification
Web robot detection
Web security
Web usage mining

ASJC Scopus subject areas

Computer Science(all)
Law

Access to Document

10.1016/j.cose.2009.05.004

Cite this

@article{2202ec488bb14f23a26dd8dbb6d4eb1c,

title = "Classification of web robots: An empirical study based on over one billion requests",

abstract = "Many studies on detection and classification of web robots have focused their attention mostly on text crawlers, and empirical experiments used relatively small data collected at universities. In this paper, we analyzed more than one billion requests to www.microsoft.com in 24 h. Web logs were made anonymous to eliminate potential privacy concerns while preserving essential characteristics (e.g., frequency, queries, etc). We have developed an effective characterization metrics, based on workload characteristics and resource types, in detecting and classifying various web robots including text crawlers, link checkers, and icon crawlers. As expected, web robot behavior was clearly different from that of typical interactive users, and different types of web robots also exhibited different characteristics. However, comparison of the similar type of web robots, text crawlers in particular, revealed different characteristics, thereby enabling characterization with reasonably high confidence level. We divided various feature metrics into five groups, and effectiveness of each group in classification is shown in polar diagram in the decreasing order of effectiveness in the clockwise direction. One can use the findings to classify likely identify of unknown web robots, and organizations can develop appropriate measures to deal with them. Our analysis is based on recent web log data collected at one of the best known site which offers truly global service. Crown",

keywords = "Web robot characterization, Web robot classification, Web robot detection, Web security, Web usage mining",

author = "Junsup Lee and Sungdeok Cha and Dongkun Lee and Hyungkyu Lee",

note = "Funding Information: Authors would like to thank Microsoft Corporation, MSRA UR in particular, for its generous support without which research reported in this paper could not have been performed. MSRA provided us raw data as well as research gift. Authors also acknowledge support provided by the following organizations: Defense Acquisition Program Administration and Agency for Defense Development; Korea SW Industry Promotion Agency through Software Engineering Technologies Development and Experts Education program; MKE (Ministry of Knowledge Economy) and IITA, Korea through ITRC (Information Technology Research Centre) support program IITA-2008-(C1090-0801-0032); MKE/MCST/IITA program 2008-S-025-02 at ETRI; and a research grant from Korea University. ",

year = "2009",

month = nov,

doi = "10.1016/j.cose.2009.05.004",

language = "English",

volume = "28",

pages = "795--802",

journal = "Computers and Security",

issn = "0167-4048",

publisher = "Elsevier Limited",

number = "8",

}

TY - JOUR

T1 - Classification of web robots

T2 - An empirical study based on over one billion requests

AU - Lee, Junsup

AU - Cha, Sungdeok

AU - Lee, Dongkun

AU - Lee, Hyungkyu

N1 - Funding Information: Authors would like to thank Microsoft Corporation, MSRA UR in particular, for its generous support without which research reported in this paper could not have been performed. MSRA provided us raw data as well as research gift. Authors also acknowledge support provided by the following organizations: Defense Acquisition Program Administration and Agency for Defense Development; Korea SW Industry Promotion Agency through Software Engineering Technologies Development and Experts Education program; MKE (Ministry of Knowledge Economy) and IITA, Korea through ITRC (Information Technology Research Centre) support program IITA-2008-(C1090-0801-0032); MKE/MCST/IITA program 2008-S-025-02 at ETRI; and a research grant from Korea University.

PY - 2009/11

Y1 - 2009/11

N2 - Many studies on detection and classification of web robots have focused their attention mostly on text crawlers, and empirical experiments used relatively small data collected at universities. In this paper, we analyzed more than one billion requests to www.microsoft.com in 24 h. Web logs were made anonymous to eliminate potential privacy concerns while preserving essential characteristics (e.g., frequency, queries, etc). We have developed an effective characterization metrics, based on workload characteristics and resource types, in detecting and classifying various web robots including text crawlers, link checkers, and icon crawlers. As expected, web robot behavior was clearly different from that of typical interactive users, and different types of web robots also exhibited different characteristics. However, comparison of the similar type of web robots, text crawlers in particular, revealed different characteristics, thereby enabling characterization with reasonably high confidence level. We divided various feature metrics into five groups, and effectiveness of each group in classification is shown in polar diagram in the decreasing order of effectiveness in the clockwise direction. One can use the findings to classify likely identify of unknown web robots, and organizations can develop appropriate measures to deal with them. Our analysis is based on recent web log data collected at one of the best known site which offers truly global service. Crown

AB - Many studies on detection and classification of web robots have focused their attention mostly on text crawlers, and empirical experiments used relatively small data collected at universities. In this paper, we analyzed more than one billion requests to www.microsoft.com in 24 h. Web logs were made anonymous to eliminate potential privacy concerns while preserving essential characteristics (e.g., frequency, queries, etc). We have developed an effective characterization metrics, based on workload characteristics and resource types, in detecting and classifying various web robots including text crawlers, link checkers, and icon crawlers. As expected, web robot behavior was clearly different from that of typical interactive users, and different types of web robots also exhibited different characteristics. However, comparison of the similar type of web robots, text crawlers in particular, revealed different characteristics, thereby enabling characterization with reasonably high confidence level. We divided various feature metrics into five groups, and effectiveness of each group in classification is shown in polar diagram in the decreasing order of effectiveness in the clockwise direction. One can use the findings to classify likely identify of unknown web robots, and organizations can develop appropriate measures to deal with them. Our analysis is based on recent web log data collected at one of the best known site which offers truly global service. Crown

KW - Web robot characterization

KW - Web robot classification

KW - Web robot detection

KW - Web security

KW - Web usage mining

UR - http://www.scopus.com/inward/record.url?scp=71849091131&partnerID=8YFLogxK

U2 - 10.1016/j.cose.2009.05.004

DO - 10.1016/j.cose.2009.05.004

M3 - Article

AN - SCOPUS:71849091131

SN - 0167-4048

VL - 28

SP - 795

EP - 802

JO - Computers and Security

JF - Computers and Security

IS - 8

ER -

Classification of web robots: An empirical study based on over one billion requests

Abstract

Bibliographical note

Keywords

ASJC Scopus subject areas

Access to Document

Other files and links

Fingerprint

Cite this