2010
DOI: 10.1007/978-3-642-12082-4_16
|View full text |Cite
|
Sign up to set email alerts
|

Shall I Trust a Recommendation? Towards an Evaluation of the Trustworthiness of Recommender Sites

Abstract: Abstract. We present a preliminary study the aim of which is to provide a high level model for the evaluation of the trustworthiness of recommender systems as e-commerce services. We identify and comment on relevant trustworthiness indicators for the following different perspectives: user interface, content information and quality of recommendations.

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
1
1
1
1

Citation Types

1
5
0

Year Published

2011
2011
2016
2016

Publication Types

Select...
3
2

Relationship

0
5

Authors

Journals

citations
Cited by 5 publications
(6 citation statements)
references
References 19 publications
1
5
0
Order By: Relevance
“…The measurements performed in this study were done using "live" controlled experiments (A/B tests) in partnership with a wellknown Italian publishing firm. One of our findings, supporting prior observations [26,18,20], is that accuracy of recommendations alone does not explain economics of purchasing behavior of customers and also does not explain how much they trust these recommendations. Therefore, we also considered diversity of recommendations, besides their accuracy, and compared performance of some of the context-aware, contentbased and random recommendations in terms of these two measures.…”
Section: Discussionsupporting
confidence: 85%
See 1 more Smart Citation
“…The measurements performed in this study were done using "live" controlled experiments (A/B tests) in partnership with a wellknown Italian publishing firm. One of our findings, supporting prior observations [26,18,20], is that accuracy of recommendations alone does not explain economics of purchasing behavior of customers and also does not explain how much they trust these recommendations. Therefore, we also considered diversity of recommendations, besides their accuracy, and compared performance of some of the context-aware, contentbased and random recommendations in terms of these two measures.…”
Section: Discussionsupporting
confidence: 85%
“…However, [18] demonstrated that the user's familiarity with the recommendations increased trust in recommender's benevolence and integrity, but not trust in its competency. [35] showed that the way familiar and unfamiliar items are balanced in a recommendation list influences users' trust in perceived usefulness of, and satisfaction with RSes [20] stated that the accuracy of the predictions provided by a RS is only one of the possible variables affecting the service overall trustworthiness. Many scholars have demonstrated that diversity can have an important role on both trust and the economics of customers' behavior.…”
Section: Prior Workmentioning
confidence: 99%
“…For a while, it was argued that trust could only exist between people but that has been challenged by research into human–computer interaction (Lenzini et al ., ). Some propose that websites can be objects of trust, thanks to the intentionality of the actors posting on them: ‘Since trust can mitigate risk, fear and complexity in the offline environment, it is likely that it can do the same in the on‐line environment’ (Corritore et al ., : 738).…”
Section: Theoretical Frameworkmentioning
confidence: 98%
“…Closely tied with credibility is the capacity for a potential booker to feel confident that they can trust the review. Accordingly, the indicators of trustworthiness include user interface, content, and quality of recommendation (Lenzini, van Houten, Huijsen, & Melenhorst, 2010).…”
Section: Tripadvisormentioning
confidence: 99%