Towards Social Choice-based Explanations in Group Recommender Systems

Thi Ngoc Trang Tran, Müslüm Atas, Alexander Felfernig, Viet Man Le, Ralph Samer, Martin Stettinger

Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

Abstract

Explanations help users to better understand why a set of items has been recommended. Compared to single user recommender systems, explanations in group recommender systems have further goals. Examples thereof are fairness which helps to take into account as much as possible group members' preferences and consensus which persuades group members to agree on a decision. This paper proposes different explanation types and investigates which explanation best helps to increase the fairness perception, consensus perception, and satisfaction of group members with regard to group recommendations. We conducted a user study to evaluate the proposed explanations. The results show that explanations which take into account preferences of all or the majority of group members achieve the best results in terms of the mentioned aspects. Moreover, there exist positive correlations among these aspects, i.e., as the perceived fairness (or the perceived consensus) of explanations increases, so does the satisfaction of users with regard to group recommendations. In addition, in the context of repeated decisions, the inclusion of group members' satisfaction from previous decisions in the explanations helps to improve the fairness perception of users with regard to group recommendations.
Originalspracheenglisch
TitelProceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization
ErscheinungsortLarnaca, Cyprus
Herausgeber (Verlag)ACM/IEEE
Seiten13-21
Seitenumfang9
ISBN (Print)978-1-4503-6021-0
DOIs
PublikationsstatusVeröffentlicht - 2019
Veranstaltung27th ACM Conference On User Modelling, Adaptation And Personalization - Larnaca, Zypern
Dauer: 9 Jun 201912 Jun 2019

Konferenz

Konferenz27th ACM Conference On User Modelling, Adaptation And Personalization
KurztitelUMAP 2019
LandZypern
OrtLarnaca
Zeitraum9/06/1912/06/19

Fingerprint

Recommender systems

Schlagwörter

    Dies zitieren

    Tran, T. N. T., Atas, M., Felfernig, A., Le, V. M., Samer, R., & Stettinger, M. (2019). Towards Social Choice-based Explanations in Group Recommender Systems. in Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization (S. 13-21). Larnaca, Cyprus: ACM/IEEE. https://doi.org/10.1145/3320435.3320437

    Towards Social Choice-based Explanations in Group Recommender Systems. / Tran, Thi Ngoc Trang; Atas, Müslüm; Felfernig, Alexander; Le, Viet Man; Samer, Ralph; Stettinger, Martin.

    Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization. Larnaca, Cyprus : ACM/IEEE, 2019. S. 13-21.

    Publikation: Beitrag in Buch/Bericht/KonferenzbandBeitrag in einem KonferenzbandForschungBegutachtung

    Tran, TNT, Atas, M, Felfernig, A, Le, VM, Samer, R & Stettinger, M 2019, Towards Social Choice-based Explanations in Group Recommender Systems. in Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization. ACM/IEEE, Larnaca, Cyprus, S. 13-21, Larnaca, Zypern, 9/06/19. https://doi.org/10.1145/3320435.3320437
    Tran TNT, Atas M, Felfernig A, Le VM, Samer R, Stettinger M. Towards Social Choice-based Explanations in Group Recommender Systems. in Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization. Larnaca, Cyprus: ACM/IEEE. 2019. S. 13-21 https://doi.org/10.1145/3320435.3320437
    Tran, Thi Ngoc Trang ; Atas, Müslüm ; Felfernig, Alexander ; Le, Viet Man ; Samer, Ralph ; Stettinger, Martin. / Towards Social Choice-based Explanations in Group Recommender Systems. Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization. Larnaca, Cyprus : ACM/IEEE, 2019. S. 13-21
    @inproceedings{050595b774884754a757f1f3fa9eb638,
    title = "Towards Social Choice-based Explanations in Group Recommender Systems",
    abstract = "Explanations help users to better understand why a set of items has been recommended. Compared to single user recommender systems, explanations in group recommender systems have further goals. Examples thereof are fairness which helps to take into account as much as possible group members' preferences and consensus which persuades group members to agree on a decision. This paper proposes different explanation types and investigates which explanation best helps to increase the fairness perception, consensus perception, and satisfaction of group members with regard to group recommendations. We conducted a user study to evaluate the proposed explanations. The results show that explanations which take into account preferences of all or the majority of group members achieve the best results in terms of the mentioned aspects. Moreover, there exist positive correlations among these aspects, i.e., as the perceived fairness (or the perceived consensus) of explanations increases, so does the satisfaction of users with regard to group recommendations. In addition, in the context of repeated decisions, the inclusion of group members' satisfaction from previous decisions in the explanations helps to improve the fairness perception of users with regard to group recommendations.",
    keywords = "consensus perception, fairness perception, group recommender systems, preference aggregation strategies, social choice theory, social choice-based explanations, textual explanations",
    author = "Tran, {Thi Ngoc Trang} and M{\"u}sl{\"u}m Atas and Alexander Felfernig and Le, {Viet Man} and Ralph Samer and Martin Stettinger",
    year = "2019",
    doi = "10.1145/3320435.3320437",
    language = "English",
    isbn = "978-1-4503-6021-0",
    pages = "13--21",
    booktitle = "Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization",
    publisher = "ACM/IEEE",

    }

    TY - GEN

    T1 - Towards Social Choice-based Explanations in Group Recommender Systems

    AU - Tran, Thi Ngoc Trang

    AU - Atas, Müslüm

    AU - Felfernig, Alexander

    AU - Le, Viet Man

    AU - Samer, Ralph

    AU - Stettinger, Martin

    PY - 2019

    Y1 - 2019

    N2 - Explanations help users to better understand why a set of items has been recommended. Compared to single user recommender systems, explanations in group recommender systems have further goals. Examples thereof are fairness which helps to take into account as much as possible group members' preferences and consensus which persuades group members to agree on a decision. This paper proposes different explanation types and investigates which explanation best helps to increase the fairness perception, consensus perception, and satisfaction of group members with regard to group recommendations. We conducted a user study to evaluate the proposed explanations. The results show that explanations which take into account preferences of all or the majority of group members achieve the best results in terms of the mentioned aspects. Moreover, there exist positive correlations among these aspects, i.e., as the perceived fairness (or the perceived consensus) of explanations increases, so does the satisfaction of users with regard to group recommendations. In addition, in the context of repeated decisions, the inclusion of group members' satisfaction from previous decisions in the explanations helps to improve the fairness perception of users with regard to group recommendations.

    AB - Explanations help users to better understand why a set of items has been recommended. Compared to single user recommender systems, explanations in group recommender systems have further goals. Examples thereof are fairness which helps to take into account as much as possible group members' preferences and consensus which persuades group members to agree on a decision. This paper proposes different explanation types and investigates which explanation best helps to increase the fairness perception, consensus perception, and satisfaction of group members with regard to group recommendations. We conducted a user study to evaluate the proposed explanations. The results show that explanations which take into account preferences of all or the majority of group members achieve the best results in terms of the mentioned aspects. Moreover, there exist positive correlations among these aspects, i.e., as the perceived fairness (or the perceived consensus) of explanations increases, so does the satisfaction of users with regard to group recommendations. In addition, in the context of repeated decisions, the inclusion of group members' satisfaction from previous decisions in the explanations helps to improve the fairness perception of users with regard to group recommendations.

    KW - consensus perception

    KW - fairness perception

    KW - group recommender systems

    KW - preference aggregation strategies

    KW - social choice theory

    KW - social choice-based explanations

    KW - textual explanations

    U2 - 10.1145/3320435.3320437

    DO - 10.1145/3320435.3320437

    M3 - Conference contribution

    SN - 978-1-4503-6021-0

    SP - 13

    EP - 21

    BT - Proceedings of the 27th ACM Conference on User Modeling, Adaptation and Personalization

    PB - ACM/IEEE

    CY - Larnaca, Cyprus

    ER -