Analysis of Aggregated Bot and Human Traffic on E-Commerce Site

Suchacka, Grażyna

doi:10.15439/2014f346

Cited by 19 publications

(16 citation statements)

References 16 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…13) checkout_suc -the number of operations corresponding to the purchase confirmation; 14) shipping_ch -the number of pages corresponding to checking information on shipping terms and cost during the checkout process;…”

Section: A Reconstruction and Characterization Of User Sessionsmentioning

confidence: 99%

“…2) Criterion II Furthermore, we can distinguish an additional group of robots based on session features impossible to occur in human sessions [14]. We assumed that a session was performed by a bot if it meets at least one of the following three conditions:…”

Section: B Marking Robot Sessionsmentioning

confidence: 99%

“…Data mining has been extensively applied to the analysis of the Web, mainly in respect of Web traffic characterization and prediction [2][3][4] and user behavior analysis and support [5][6][7][8][9]. Some studies focused on exploring differences between the Web traffic generated by human users and robots as regards to traffic patterns and the consumption of server resources [10][11][12][13][14]. These differences are significant enough to be used in classification models and methods for robot detection.…”

Section: Introductionmentioning

confidence: 99%

“…In this paper, we use the results of our earlier study on the differences in bots' and humans' Web traffic in a Web store [14] to construct a 21-element feature vector characterizing a user session. Then we build and verify a session classification model using the Bayesian approach.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Detection of Internet robots using a Bayesian approach

Suchacka

Sobkow

2015

2015 IEEE 2nd International Conference on Cybernetics (CYBCONF)

View full text Add to dashboard Cite

A large part of Web traffic on e-commerce sites is generated not by human users but by Internet robots: search engine crawlers, shopping bots, hacking bots, etc. In practice, not all robots, especially the malicious ones, disclose their identities to a Web server and thus there is a need to develop methods for their detection and identification. This paper proposes the application of a Bayesian approach to robot detection based on characteristics of user sessions. The method is applied to the Web traffic from a real e-commerce site. Results show that the classification model based on the cluster analysis with the Ward's method and the weighted Euclidean metric is very effective in robot detection, even obtaining accuracy of above 90%.

show abstract

Section: A Reconstruction and Characterization Of User Sessionsmentioning

confidence: 99%

Section: B Marking Robot Sessionsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Detection of Internet robots using a Bayesian approach

Suchacka

Sobkow

2015

2015 IEEE 2nd International Conference on Cybernetics (CYBCONF)

View full text Add to dashboard Cite

show abstract

“…The case of the technical articles wholesale point for exploring needs of business customers. To better understand these needs we should distinguish data stored in logs, generated by bots from the traffic generated by business users [27]. Well-identified needs are the foundation for further good relationships with customers.…”

Section: Introductionmentioning

confidence: 99%

Searching for information and making purchase decisions in b2b online stores. The case of the technical articles wholesale

Wiechetek

Pawłowski²

2016

Annals of Computer Science and Information Systems

View full text Add to dashboard Cite

Abstract-This paper aims to extend current research in the area of on-line business-to-business clients preferences. A quantitative research in a form of transaction log analysis (TLA) performed by the authors allow to conclude that search log and shop basket logs are sources of viable information about business customers. Transaction log analysis allows to identify different customer groups, such as searchers and buyers. It also allows for better customization of cloud of tags, so that the results are better tailored to customers' preferences. It turns out that a large list of products (result of search process) is not a deterrent to purchasing on-line. However, in order to facilitate the purchasing process b2b platform should offer additional filtering mechanisms.The main results of the research could be used by managers of professional e-commerce b2b platforms to better understand customers' needs and develop strategies to build long lasting partnerships.Research limitations. The authors examined single b2b platform that may limit the findings generalizability, so the future research of other b2b platforms should be explored to validate the results.Our future research should be performed to obtain the interplay between quantitative and qualitative methods of b2b clients' analysis and online shopping. It will allow for better understanding of customers' engagement and clients' unique organizational culture that may have an impact on purchasing decisions and building long time relationship.

show abstract