Explaining Social Recommendations to Casual Users: Design Principles and Opportunities
C. Tsai, and P. Brusilovsky. Proceedings of the 23rd International Conference on Intelligent User Interfaces Companion, page 59:1--59:2. New York, NY, USA, ACM, (2018)
DOI: 10.1145/3180308.3180368
Abstract
Recommender systems have become popular in recent years, and ordinary users are more likely to rely on such service when completing various daily tasks. The need to design and build explainable recommender interfaces is increasing rapidly. Most of the designs of such explanations are intended to reflect the underlying algorithms by which the recommendations are computed. These approaches have been shown to be useful for obtaining system transparency and trust. However, little is known about how to design explanation interfaces for causal (non-expert) users to achieve different explanatory goals. As a first step toward understanding the user interface design factors, we conducted an international (across 13 countries) online survey of 14 active users of a social recommender system. This study captures user feedback in the field and frames it in terms of design principles and opportunities.
%0 Conference Paper
%1 Tsai:2018:ESR:3180308.3180368
%A Tsai, Chun-Hua
%A Brusilovsky, Peter
%B Proceedings of the 23rd International Conference on Intelligent User Interfaces Companion
%C New York, NY, USA
%D 2018
%I ACM
%K explanation myown recommender social-recommendation
%P 59:1--59:2
%R 10.1145/3180308.3180368
%T Explaining Social Recommendations to Casual Users: Design Principles and Opportunities
%U http://doi.acm.org/10.1145/3180308.3180368
%X Recommender systems have become popular in recent years, and ordinary users are more likely to rely on such service when completing various daily tasks. The need to design and build explainable recommender interfaces is increasing rapidly. Most of the designs of such explanations are intended to reflect the underlying algorithms by which the recommendations are computed. These approaches have been shown to be useful for obtaining system transparency and trust. However, little is known about how to design explanation interfaces for causal (non-expert) users to achieve different explanatory goals. As a first step toward understanding the user interface design factors, we conducted an international (across 13 countries) online survey of 14 active users of a social recommender system. This study captures user feedback in the field and frames it in terms of design principles and opportunities.
%@ 978-1-4503-5571-1
@inproceedings{Tsai:2018:ESR:3180308.3180368,
abstract = {Recommender systems have become popular in recent years, and ordinary users are more likely to rely on such service when completing various daily tasks. The need to design and build explainable recommender interfaces is increasing rapidly. Most of the designs of such explanations are intended to reflect the underlying algorithms by which the recommendations are computed. These approaches have been shown to be useful for obtaining system transparency and trust. However, little is known about how to design explanation interfaces for causal (non-expert) users to achieve different explanatory goals. As a first step toward understanding the user interface design factors, we conducted an international (across 13 countries) online survey of 14 active users of a social recommender system. This study captures user feedback in the field and frames it in terms of design principles and opportunities.},
acmid = {3180368},
added-at = {2018-07-25T16:30:00.000+0200},
address = {New York, NY, USA},
articleno = {59},
author = {Tsai, Chun-Hua and Brusilovsky, Peter},
biburl = {https://www.bibsonomy.org/bibtex/27f656b2c4ec5d261d0627ad8d8ed651e/brusilovsky},
booktitle = {Proceedings of the 23rd International Conference on Intelligent User Interfaces Companion},
description = {Explaining Social Recommendations to Casual Users},
doi = {10.1145/3180308.3180368},
interhash = {643c6e22fa413ea43e83aec6de424bf6},
intrahash = {7f656b2c4ec5d261d0627ad8d8ed651e},
isbn = {978-1-4503-5571-1},
keywords = {explanation myown recommender social-recommendation},
location = {Tokyo, Japan},
numpages = {2},
pages = {59:1--59:2},
publisher = {ACM},
series = {IUI '18 Companion},
timestamp = {2018-07-25T16:30:00.000+0200},
title = {Explaining Social Recommendations to Casual Users: Design Principles and Opportunities},
url = {http://doi.acm.org/10.1145/3180308.3180368},
year = 2018
}