An Analysis of Group Recommendation Heuristics for High- and Low-Involvement Items

Alexander Felfernig, Müslüm Atas, Trang Tran, Martin Stettinger, Seda Polat Erdeniz, Gerhard Leitner

Research output: Contribution to conferencePaperResearchpeer-review

Abstract

Group recommender systems are based on aggregation heuristics that help to determine a recommendation for a group. These heuristics aggregate the preferences of individual users in order to reflect the preferences of the whole group. There exist a couple of different aggregation heuristics (e.g., most pleasure, least misery, and average voting) that are applied in group recommendation scenarios. However, to some extent it is still unclear which heuristics should be applied in which context. In this paper, we analyze the impact of the item domain (low involvement vs. high involvement) on the appropriateness of aggregation heuristics (we use restaurants as an example of low-involvement items and shared apartments as an example of high-involvement ones). The results of our study show that aggregation heuristics in group recommendation should be tailored to the underlying item domain.
Translated title of the contributionAn Analysis of Group Recommendation Heuristics for High- and Low-Involvement Items
Original languageEnglish
Pages335-344
Number of pages10
Publication statusPublished - 15 Aug 2017

Fingerprint

heuristics
aggregation
Group
apartment
voting
scenario

Keywords

  • Recommender systems
  • Group decision making
  • Group recommendation Decision heuristics

Cite this

An Analysis of Group Recommendation Heuristics for High- and Low-Involvement Items. / Felfernig, Alexander; Atas, Müslüm; Tran, Trang; Stettinger, Martin; Polat Erdeniz, Seda; Leitner, Gerhard.

2017. 335-344.

Research output: Contribution to conferencePaperResearchpeer-review

@conference{63ee5fdc19e441c590eff035f9ffd7c2,
title = "An Analysis of Group Recommendation Heuristics for High- and Low-Involvement Items",
abstract = "Group recommender systems are based on aggregation heuristics that help to determine a recommendation for a group. These heuristics aggregate the preferences of individual users in order to reflect the preferences of the whole group. There exist a couple of different aggregation heuristics (e.g., most pleasure, least misery, and average voting) that are applied in group recommendation scenarios. However, to some extent it is still unclear which heuristics should be applied in which context. In this paper, we analyze the impact of the item domain (low involvement vs. high involvement) on the appropriateness of aggregation heuristics (we use restaurants as an example of low-involvement items and shared apartments as an example of high-involvement ones). The results of our study show that aggregation heuristics in group recommendation should be tailored to the underlying item domain.",
keywords = "Recommender systems, Group decision making, Group recommendation Decision heuristics",
author = "Alexander Felfernig and M{\"u}sl{\"u}m Atas and Trang Tran and Martin Stettinger and {Polat Erdeniz}, Seda and Gerhard Leitner",
year = "2017",
month = "8",
day = "15",
language = "English",
pages = "335--344",

}

TY - CONF

T1 - An Analysis of Group Recommendation Heuristics for High- and Low-Involvement Items

AU - Felfernig, Alexander

AU - Atas, Müslüm

AU - Tran, Trang

AU - Stettinger, Martin

AU - Polat Erdeniz, Seda

AU - Leitner, Gerhard

PY - 2017/8/15

Y1 - 2017/8/15

N2 - Group recommender systems are based on aggregation heuristics that help to determine a recommendation for a group. These heuristics aggregate the preferences of individual users in order to reflect the preferences of the whole group. There exist a couple of different aggregation heuristics (e.g., most pleasure, least misery, and average voting) that are applied in group recommendation scenarios. However, to some extent it is still unclear which heuristics should be applied in which context. In this paper, we analyze the impact of the item domain (low involvement vs. high involvement) on the appropriateness of aggregation heuristics (we use restaurants as an example of low-involvement items and shared apartments as an example of high-involvement ones). The results of our study show that aggregation heuristics in group recommendation should be tailored to the underlying item domain.

AB - Group recommender systems are based on aggregation heuristics that help to determine a recommendation for a group. These heuristics aggregate the preferences of individual users in order to reflect the preferences of the whole group. There exist a couple of different aggregation heuristics (e.g., most pleasure, least misery, and average voting) that are applied in group recommendation scenarios. However, to some extent it is still unclear which heuristics should be applied in which context. In this paper, we analyze the impact of the item domain (low involvement vs. high involvement) on the appropriateness of aggregation heuristics (we use restaurants as an example of low-involvement items and shared apartments as an example of high-involvement ones). The results of our study show that aggregation heuristics in group recommendation should be tailored to the underlying item domain.

KW - Recommender systems

KW - Group decision making

KW - Group recommendation Decision heuristics

M3 - Paper

SP - 335

EP - 344

ER -