ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering

Denis Kotkov; Zhao Qian; Launis Kati; Mats Neovius

doi:10.1145/3383313.3412221

ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering

Denis Kotkov, Zhao Qian, Launis Kati, Mats Neovius

Information Technology

Research output: Chapter in Book/Conference proceeding › Conference contribution › Scientific › peer-review

4 Downloads (Pure)

Abstract

Related item recommendations have a long history in recommender systems, but they tend to be a static list of similar items with respect to a target item of interest without any support of user control. In this paper, we propose ClusterExplorer, a novel approach for enabling user control over related recommendations. The approach allows users to explore the latent space of user-item interactions through controlling related recommendations. We evaluated ClusterExplorer in the book domain with 42 participants recruited in a public library and found that our approach has higher user satisfaction of browsing items and is more helpful in finding interesting items compared to traditional related item recommendations.

Original language	English
Title of host publication	RecSys '20: Fourteenth ACM Conference on Recommender Systems
Publisher	Association for Computing Machinery
Pages	432–437
Number of pages	6
ISBN (Electronic)	9781450375832
ISBN (Print)	9781450375832
DOIs	https://doi.org/10.1145/3383313.3412221
Publication status	Published - 22 Sep 2020
MoE publication type	A4 Article in a conference publication
Event	conference; 2020-09-22; 2020-09-26 - Fourteenth ACM Conference on Recommender Systems Duration: 22 Sep 2020 → 26 Sep 2020

Conference

Conference	conference; 2020-09-22; 2020-09-26
Period	22/09/20 → 26/09/20

Keywords

conversational recommender systems
critiquing recommender systems
information exploration tool
interactive recommendation
recommender systems
related item recommendations
user control
user interfaces

Access to Document

10.1145/3383313.3412221

!ClusterExplorer Enable User Control over Related Recommendations via Collaborative Filtering and Clustering.pdfFinal published version, 0.99 MBLicence: CC BY

Fingerprint Dive into the research topics of 'ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering'. Together they form a unique fingerprint.

Cite this

@inproceedings{3390358f54944bdabaee2e9f88c9dff4,

title = "ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering",

abstract = "Related item recommendations have a long history in recommender systems, but they tend to be a static list of similar items with respect to a target item of interest without any support of user control. In this paper, we propose ClusterExplorer, a novel approach for enabling user control over related recommendations. The approach allows users to explore the latent space of user-item interactions through controlling related recommendations. We evaluated ClusterExplorer in the book domain with 42 participants recruited in a public library and found that our approach has higher user satisfaction of browsing items and is more helpful in finding interesting items compared to traditional related item recommendations.",

keywords = "conversational recommender systems, critiquing recommender systems, information exploration tool, interactive recommendation, recommender systems, related item recommendations, user control, user interfaces, conversational recommender systems, critiquing recommender systems, information exploration tool, interactive recommendation, recommender systems, related item recommendations, user control, user interfaces, conversational recommender systems, critiquing recommender systems, information exploration tool, interactive recommendation, recommender systems, related item recommendations, user control, user interfaces",

author = "Denis Kotkov and Zhao Qian and Launis Kati and Mats Neovius",

year = "2020",

month = sep,

day = "22",

doi = "10.1145/3383313.3412221",

language = "English",

isbn = "9781450375832",

pages = "432–437",

booktitle = "RecSys '20: Fourteenth ACM Conference on Recommender Systems",

publisher = "Association for Computing Machinery",

note = "conference; 2020-09-22; 2020-09-26 ; Conference date: 22-09-2020 Through 26-09-2020",

}

ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering. / Kotkov, Denis; Qian, Zhao; Kati, Launis; Neovius, Mats.

RecSys '20: Fourteenth ACM Conference on Recommender Systems. Association for Computing Machinery, 2020. p. 432–437.

Research output: Chapter in Book/Conference proceeding › Conference contribution › Scientific › peer-review

TY - GEN

T1 - ClusterExplorer: Enable User Control over Related Recommendations via Collaborative Filtering and Clustering

AU - Kotkov, Denis

AU - Qian, Zhao

AU - Kati, Launis

AU - Neovius, Mats

PY - 2020/9/22

Y1 - 2020/9/22

N2 - Related item recommendations have a long history in recommender systems, but they tend to be a static list of similar items with respect to a target item of interest without any support of user control. In this paper, we propose ClusterExplorer, a novel approach for enabling user control over related recommendations. The approach allows users to explore the latent space of user-item interactions through controlling related recommendations. We evaluated ClusterExplorer in the book domain with 42 participants recruited in a public library and found that our approach has higher user satisfaction of browsing items and is more helpful in finding interesting items compared to traditional related item recommendations.

AB - Related item recommendations have a long history in recommender systems, but they tend to be a static list of similar items with respect to a target item of interest without any support of user control. In this paper, we propose ClusterExplorer, a novel approach for enabling user control over related recommendations. The approach allows users to explore the latent space of user-item interactions through controlling related recommendations. We evaluated ClusterExplorer in the book domain with 42 participants recruited in a public library and found that our approach has higher user satisfaction of browsing items and is more helpful in finding interesting items compared to traditional related item recommendations.

KW - conversational recommender systems

KW - critiquing recommender systems

KW - information exploration tool

KW - interactive recommendation

KW - recommender systems

KW - related item recommendations

KW - user control

KW - user interfaces

KW - conversational recommender systems

KW - critiquing recommender systems

KW - information exploration tool

KW - interactive recommendation

KW - recommender systems

KW - related item recommendations

KW - user control

KW - user interfaces

KW - conversational recommender systems

KW - critiquing recommender systems

KW - information exploration tool

KW - interactive recommendation

KW - recommender systems

KW - related item recommendations

KW - user control

KW - user interfaces

UR - http://www.scopus.com/inward/record.url?scp=85092747265&partnerID=8YFLogxK

U2 - 10.1145/3383313.3412221

DO - 10.1145/3383313.3412221

M3 - Conference contribution

SN - 9781450375832

SP - 432

EP - 437

BT - RecSys '20: Fourteenth ACM Conference on Recommender Systems

PB - Association for Computing Machinery

T2 - conference; 2020-09-22; 2020-09-26

Y2 - 22 September 2020 through 26 September 2020

ER -