作者: A.P.W. Eliëns , Y. Wang
DOI:
关键词:
摘要: In this paper we propose a tentative framework (R3) for adapting sequence of predictions (guided tour) generated by what call serial recommender. The R3 (rate, recommend, regret) is applied to the construction personalized guided tours, based on expert advice, in domain cultural heritage, particular digital dossiers about contemporary art. Guided tours are first instance obtained tracking users. Our proposal variant decision theory, that uses regret function measure difference between proposed and finite collection decisions. our framework, personalization may then be seen as minimization problem over weighting scheme, expressing relative importance experts which available. aim arrive at formalization recommendation sequences tours) allows adaptation individual user preferences revision weight attached advice feedback.