Comments
Reference InformationTitle: Tagsplanations: explaining recommendations using tags
Authors: Jesse Vig, Shilad Sen, John Riedl
Presentation Venue: IUI 2009: Proceedings of the 14th international conference on Intelligent user interfaces; February 8-11, 2009; Sanibel Island, Florida, USA
Authors: Jesse Vig, Shilad Sen, John Riedl
Presentation Venue: IUI 2009: Proceedings of the 14th international conference on Intelligent user interfaces; February 8-11, 2009; Sanibel Island, Florida, USA
Summary
In this paper, the authors talk about Tagsplanations: a design implemented on a movie recommender website. Not only are recommendations made to users but, these recommendations are also in turn explained for(i.e.: Why was a certain recommendation given). They discuss an algorithm used to determine tag relevance and tag preference and present a study to see how users like tagsplanations. If an item is similar in feature, for example, the relevance would be high, but if the user preferred a certain type of item or brand, the preference would be high.
To test the implementation of tagsplanations, the researchers used MovieLens, a movie recommendation site. They computed tag preference based on user behavior (by movie ratings in their MovieLens implementation).
An experiment was also conducted in order to see how users responded to the different types of recommendations. The conclusions were that tag preference was more important than tag relevance. Concerning effectiveness, the two were about the same. For mood compatibility, relevance was rated higher.
Discussion
In this paper, the authors talk about Tagsplanations: a design implemented on a movie recommender website. Not only are recommendations made to users but, these recommendations are also in turn explained for(i.e.: Why was a certain recommendation given). They discuss an algorithm used to determine tag relevance and tag preference and present a study to see how users like tagsplanations. If an item is similar in feature, for example, the relevance would be high, but if the user preferred a certain type of item or brand, the preference would be high.
To test the implementation of tagsplanations, the researchers used MovieLens, a movie recommendation site. They computed tag preference based on user behavior (by movie ratings in their MovieLens implementation).
An experiment was also conducted in order to see how users responded to the different types of recommendations. The conclusions were that tag preference was more important than tag relevance. Concerning effectiveness, the two were about the same. For mood compatibility, relevance was rated higher.
Discussion
I thought this paper was interesting and relevant. In looking at their interface, I think they’ve done a good job presenting their information in a concise and straightforward way.
The research reminded me of recommendation systems like Amazon, and Google. I can easily see this design as benefiting to users. The authors explained their motivations well and provided an in-depth user study. For future works the researchers mentioned possibly making a way in which users can inform the system when it is wrong and the system can inform users when different actions impact the results they receive. They also mentioned exploring other techniques to estimate tag preference.
No comments:
Post a Comment