Client- and server-side revisitation prediction with SUPRA

Papadakis, G; Kawase, R; Herder, E

dc.contributor.author	Papadakis, G	en
dc.contributor.author	Kawase, R	en
dc.contributor.author	Herder, E	en
dc.date.accessioned	2014-03-01T02:53:34Z
dc.date.available	2014-03-01T02:53:34Z
dc.date.issued	2012	en
dc.identifier.uri	https://dspace.lib.ntua.gr/xmlui/handle/123456789/36423
dc.subject	Contextual support	en
dc.subject	Revisitation prediction	en
dc.subject	Web behavior	en
dc.subject.other	Client-side application	en
dc.subject.other	Collaborative application	en
dc.subject.other	Contextual information	en
dc.subject.other	Contextual support	en
dc.subject.other	Generic frameworks	en
dc.subject.other	Real-world datasets	en
dc.subject.other	Revisitation	en
dc.subject.other	Training data	en
dc.subject.other	Web behavior	en
dc.subject.other	Data processing	en
dc.subject.other	Learning algorithms	en
dc.subject.other	Semantic Web	en
dc.subject.other	Semantics	en
dc.subject.other	Forecasting	en
dc.title	Client- and server-side revisitation prediction with SUPRA	en
heal.type	conferenceItem	en
heal.identifier.primary	10.1145/2254129.2254149	en
heal.identifier.secondary	http://dx.doi.org/10.1145/2254129.2254149	en
heal.identifier.secondary	14	en
heal.publicationDate	2012	en
heal.abstract	Users of collaborative applications as well as individual users in their private environment return to previously visited Web pages for various reasons; apart from pages visited due to backtracking, they typically have a number of favorite or important pages that they monitor or tasks that reoccur on an infrequent basis. In this paper, we introduce a library of methods that facilitate revisitation through the effective prediction of the next page request. It is based on a generic framework that inherently incorporates contextual information, handling uniformly both server- and the client-side applications. Unlike other existing approaches, the methods it encompasses are real-time, since they do not rely on training data or machine learning algorithms. We evaluate them over two large, real-world datasets, with the outcomes suggesting a significant improvement over methods typically used in this context. We have also made our implementation and data publicly available, thus encouraging other researchers to use it as a benchmark and to extend it with new techniques for supporting user's navigational activity. Copyright 2012 ACM.	en
heal.journalName	ACM International Conference Proceeding Series	en
dc.identifier.doi	10.1145/2254129.2254149	en