Maximal paths recipe for constructing Web user sessions

Bayır, Murat Ali; Toroslu, İsmail Hakkı

doi:10.1007/s11280-022-01024-3

Cited by 3 publications

(1 citation statement)

References 29 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The session identification step is responsible to identify the user's navigational behavior. Here, the user's session is identified by the Time gap heuristics in which a set of pages is considered as a single user session if those pages are requested within a specified period, which is 30 minutes and if the request was given after that threshold, it is considered as a new request (session) (Bayir & Toroslu, 2022). At the end of this layer, the data became ready to be used in the pattern discovery phase.…”

Section: Session Identificationmentioning

confidence: 99%

Web Mining Techniques - A Framework to Enhance Customer Retention

Ouf

Helmy

Ashraf

2023

International Journal of E-Collaboration

View full text Add to dashboard Cite

In e-commerce, retaining customers on the web is a difficult task that requires a good understanding of customers' behavior to be able to predict their needs and interests. Web usage mining (WUM), which is the application of data mining techniques to improve business, helps in understanding customers' behavior on the web. Therefore, this paper proposes and implements a framework to enhance the quality of customer recommendations. Providing customers with what they are looking for helps increase their satisfaction, which will lead to improved retention with the company. The proposed framework was tested and evaluated. The result of testing the proposed framework illustrates that the recommendations based on merged techniques (like clustering, classification, association, and sequential discovery) achieve strong accuracy with a precision value of 74%, coverage of 100%, and an average overall efficiency of F-measure of 86%. which means that the merged technique outperformed each technique and attained much higher overall coverage.

show abstract

Section: Session Identificationmentioning

confidence: 99%