No abstract

No abstract

No abstract

No abstract

Abstract:RecommenderS ystems ares oftwaret oolsa nd techniques providing suggestions foritems to be of usetoauser.Recommendersystems have proven to be av aluablem eans foro nlineu sers to cope with thev irtual information overload and have become oneo ft he most powerful andp opular toolsi n electronicc ommerce. Correspondingly,v arious techniquesf or recommendation generation have been proposed during thel astd ecade. In this paperw ep resent an ew benchmark framework. It allows researcherso rpractitioners to quickly tryo ut and compare different recommendation methodson newdatasets. Extendingthe framework is easy thanks to as implea nd well-defined Application Programming Interface( API).I t contains ap lug-in mechanisma llowingo therst od evelop theiro wn algorithms and incorporatethemin theframework.A n interactivegraphical user interfaceisprovided fors etting newb enchmarks, integraten ew plug-ins with thef ramework,s etting up configurations andexploring benchmarkresults.