Recommender Systems Evaluation

Bellogín, Alejandro; Said, Alan

doi:10.1007/978-1-4614-7163-9_110162-1

Cited by 2 publications

(2 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…User and online studies do not have that flexibility: only a few alternative RSs can be compared, by letting the users to try them, either in a controlled situation or in the wild (Bellogín and Said 2018;Gunawardana and Shani 2015;Knijnenburg et al 2012;Pu, Chen, and Hu 2011). Conversely, in user and online studies, the collected user/system interactions can be analzsed and the users' reactions even to "novel" recommendations can be observed.…”

Section: Evaluating Tourist Rssmentioning

confidence: 99%

See 1 more Smart Citation

Building effective recommender systems for tourists

Massimo

Ricci

2022

AI Magazine

View full text Add to dashboard Cite

Recommender systems (RSs) are personalized information search and discovery applications helping users to identify and choose useful items and information. In this paper, we focus on the tourism application scenario and its specific requirements. We discuss a novel RS approach that copes with the specific application constraints of the domain and produces recommendations that better match the true needs of tourists. We illustrate the proposed next POI recommendation approach in a case study and we compare it with a state-of-the-art nearest neighbor-based next item RS. With the analysis of this case study, we aim at illustrating the specific features of the compared approaches also with the goal to raise the discussion on RSs validation methods, with a particular attention to tourism applications. We finally discuss some significant limitations of current evaluation approaches that must be addressed in future studies.

show abstract

Section: Evaluating Tourist Rssmentioning

confidence: 99%