Abstract
Most recommender systems propose items to individual users. However, in domains such as tourism, people often consume items in groups rather than individually. Different individual preferences in such a group can be difficult to resolve, and often compromises need to be made. Social choice strategies can be used to aggregate the preferences of individuals. We evaluated two explainable modified preference aggregation strategies in a between-subject study (n=200), and compared them with two baseline strategies for groups that are also explainable, in two scenarios: high divergence (group members with different travel preferences) and low divergence (group members with similar travel preferences). Generally, all investigated aggregation strategies performed well in terms of perceived individual and group satisfaction and perceived fairness. The results also indicate that participants were sensitive to a dictator-based strategy, which affected both their individual and group satisfaction negatively (compared to the other strategies).
Original language | English |
---|---|
Title of host publication | Proceedings of the 31st ACM Conference on Hypertext and Social Media (HT ’20), |
Pages | 187–196 |
Number of pages | 10 |
ISBN (Electronic) | 978-1-4503-7098-1 |
DOIs | |
Publication status | Published - 2020 |
Event | HT'20: 31st ACM Conference on Hypertext and Social Media - Online event, United States Duration: 13 Jul 2020 → 15 Jul 2020 Conference number: 31 |
Conference
Conference | HT'20: 31st ACM Conference on Hypertext and Social Media |
---|---|
Abbreviated title | HT'20 |
Country/Territory | United States |
City | Online event |
Period | 13/07/20 → 15/07/20 |
Other | Virtual/online event due to COVID-19 |
Keywords
- Explainable aggregation strategies
- Group recommendation
- Human-centered computing user studies