Collection recommender systems suggest groups of items that work well as a whole. The interaction effects between items is an important consideration, but the vast space of possible collections makes it difficult to analyze. In this paper, we present a class of games with a purpose for building collections where users create collections and, using an output agreement model, they are awarded points based on the collections that match. The data from these games will help researchers develop guidelines for collection recommender systems among other applications. We conducted a pilot study of the game prototype which indicated that it was fun and challenging for users, and that the data obtained had the characteristics necessary to gain insights into the interaction effects among items. We present the game and these results followed by a discussion of the next steps necessary to bring games to bear on the problem of creating harmonious groups.
[1]
Scott P. Robertson,et al.
Proceedings of the SIGCHI Conference on Human Factors in Computing Systems
,
1991
.
[2]
Laura A. Dabbish,et al.
Designing games with a purpose
,
2008,
CACM.
[3]
Manuel Blum,et al.
Peekaboom: a game for locating objects in images
,
2006,
CHI.
[4]
Edith Law,et al.
Input-agreement: a new mechanism for collecting data using human computation games
,
2009,
CHI.
[5]
Derek L. Hansen,et al.
Mixing it up: recommending collections of items
,
2009,
CHI.
[6]
Luis von Ahn,et al.
Matchin: eliciting user preferences with an online game
,
2009,
CHI.