FairRecKit: A Web-based analysis software for recommender evaluations

Christine Bauer; Lennard Chung; Aleksej Cornelissen; Isabelle van Driessel; Diede van der Hoorn; Yme de Jong; Lan Le; Sanaz Najiyan Tabriz; Roderick Spaans; Casper Thijsen; Robert Verbeeten; Vos Wesseling; Fern Wieland

doi:10.1145/3576840.3578274

FairRecKit: A Web-based analysis software for recommender evaluations

Christine Bauer^*, Lennard Chung, Aleksej Cornelissen, Isabelle van Driessel, Diede van der Hoorn, Yme de Jong, Lan Le, Sanaz Najiyan Tabriz, Roderick Spaans, Casper Thijsen, Robert Verbeeten, Vos Wesseling, Fern Wieland

^*Corresponding author for this work

Research output: Contribution to conference › Paper › Academic

Abstract

FairRecKit is a web-based analysis software that supports researchers in performing, analyzing, and understanding recommendation computations. The idea behind FairRecKit is to facilitate the in-depth analysis of recommendation outcomes considering fairness aspects. With (nested) filters on user or item attributes, metrics can easily be compared across user and item subgroups. Further, (nested) filters can be used on the dataset level; this way, recommendation outcomes can be compared across several sub-datasets to analyze for differences considering fairness aspects. The software currently features five datasets, 11 metrics, and 21 recommendation algorithms to be used in computational experimentation. It is open source and developed in a modular manner to facilitate extension. The analysis software consists of two components: A software package (FairRecKitLib) for running recommendation algorithms on the available datasets and a web-based user interface (FairRecKitApp) to start experiments, retrieve results of previous experiments, and analyze details. The application also comes with extensive documentation and options for result customization, which makes for a flexible tool that supports in-depth analysis.

Original language	English
Pages	438-443
Number of pages	6
DOIs	https://doi.org/10.1145/3576840.3578274
Publication status	Published - 19 Mar 2023
Event	8th ACM SIGIR Conference on Human Information Interaction and Retrieval - Austin, United States Duration: 19 Mar 2023 → 23 Mar 2023 Conference number: 8 http://sigir.org/chiir2023/

Conference

Conference	8th ACM SIGIR Conference on Human Information Interaction and Retrieval
Abbreviated title	CHIIR 2023
Country/Territory	United States
City	Austin
Period	19/03/23 → 23/03/23
Internet address	http://sigir.org/chiir2023/

Bibliographical note

Funding Information:
FairRecKit has been developed by students within the Software Project course of the Bachelor program Computer Science at Utrecht University, commissioned by Christine Bauer. The project was funded by the Department of Information and Computing Sciences of Utrecht University.

Publisher Copyright:
© 2023 Owner/Author.

Keywords

FairRecKit
resource
toolkit
analysis software
web-based evaluation
analysis
recommender systems
music
movies
evaluation
web-based

Access to Document

10.1145/3576840.3578274

Cite this

Bauer, C., Chung, L., Cornelissen, A., van Driessel, I., van der Hoorn, D., de Jong, Y., Le, L., Najiyan Tabriz, S., Spaans, R., Thijsen, C., Verbeeten, R., Wesseling, V., & Wieland, F. (2023). FairRecKit: A Web-based analysis software for recommender evaluations. 438-443. Paper presented at 8th ACM SIGIR Conference on Human Information Interaction and Retrieval, Austin, Texas, United States. https://doi.org/10.1145/3576840.3578274

@conference{4540046fbc384c8bac35b0fec4960bbf,

title = "FairRecKit: A Web-based analysis software for recommender evaluations",

abstract = "FairRecKit is a web-based analysis software that supports researchers in performing, analyzing, and understanding recommendation computations. The idea behind FairRecKit is to facilitate the in-depth analysis of recommendation outcomes considering fairness aspects. With (nested) filters on user or item attributes, metrics can easily be compared across user and item subgroups. Further, (nested) filters can be used on the dataset level; this way, recommendation outcomes can be compared across several sub-datasets to analyze for differences considering fairness aspects. The software currently features five datasets, 11 metrics, and 21 recommendation algorithms to be used in computational experimentation. It is open source and developed in a modular manner to facilitate extension. The analysis software consists of two components: A software package (FairRecKitLib) for running recommendation algorithms on the available datasets and a web-based user interface (FairRecKitApp) to start experiments, retrieve results of previous experiments, and analyze details. The application also comes with extensive documentation and options for result customization, which makes for a flexible tool that supports in-depth analysis.",

keywords = "FairRecKit, resource, toolkit, analysis software, web-based evaluation, analysis, recommender systems, music, movies, evaluation, web-based",

author = "Christine Bauer and Lennard Chung and Aleksej Cornelissen and {van Driessel}, Isabelle and {van der Hoorn}, Diede and {de Jong}, Yme and Lan Le and {Najiyan Tabriz}, Sanaz and Roderick Spaans and Casper Thijsen and Robert Verbeeten and Vos Wesseling and Fern Wieland",

note = "Funding Information: FairRecKit has been developed by students within the Software Project course of the Bachelor program Computer Science at Utrecht University, commissioned by Christine Bauer. The project was funded by the Department of Information and Computing Sciences of Utrecht University. Publisher Copyright: {\textcopyright} 2023 Owner/Author.; 8th ACM SIGIR Conference on Human Information Interaction and Retrieval, CHIIR 2023 ; Conference date: 19-03-2023 Through 23-03-2023",

year = "2023",

month = mar,

day = "19",

doi = "10.1145/3576840.3578274",

language = "English",

pages = "438--443",

url = "http://sigir.org/chiir2023/",

}

Bauer, C, Chung, L, Cornelissen, A, van Driessel, I, van der Hoorn, D, de Jong, Y, Le, L, Najiyan Tabriz, S, Spaans, R, Thijsen, C, Verbeeten, R, Wesseling, V & Wieland, F 2023, 'FairRecKit: A Web-based analysis software for recommender evaluations', Paper presented at 8th ACM SIGIR Conference on Human Information Interaction and Retrieval, Austin, United States, 19/03/23 - 23/03/23 pp. 438-443. https://doi.org/10.1145/3576840.3578274

TY - CONF

T1 - FairRecKit

T2 - 8th ACM SIGIR Conference on Human Information Interaction and Retrieval

AU - Bauer, Christine

AU - Chung, Lennard

AU - Cornelissen, Aleksej

AU - van Driessel, Isabelle

AU - van der Hoorn, Diede

AU - de Jong, Yme

AU - Le, Lan

AU - Najiyan Tabriz, Sanaz

AU - Spaans, Roderick

AU - Thijsen, Casper

AU - Verbeeten, Robert

AU - Wesseling, Vos

AU - Wieland, Fern

N1 - Conference code: 8

PY - 2023/3/19

Y1 - 2023/3/19

N2 - FairRecKit is a web-based analysis software that supports researchers in performing, analyzing, and understanding recommendation computations. The idea behind FairRecKit is to facilitate the in-depth analysis of recommendation outcomes considering fairness aspects. With (nested) filters on user or item attributes, metrics can easily be compared across user and item subgroups. Further, (nested) filters can be used on the dataset level; this way, recommendation outcomes can be compared across several sub-datasets to analyze for differences considering fairness aspects. The software currently features five datasets, 11 metrics, and 21 recommendation algorithms to be used in computational experimentation. It is open source and developed in a modular manner to facilitate extension. The analysis software consists of two components: A software package (FairRecKitLib) for running recommendation algorithms on the available datasets and a web-based user interface (FairRecKitApp) to start experiments, retrieve results of previous experiments, and analyze details. The application also comes with extensive documentation and options for result customization, which makes for a flexible tool that supports in-depth analysis.

AB - FairRecKit is a web-based analysis software that supports researchers in performing, analyzing, and understanding recommendation computations. The idea behind FairRecKit is to facilitate the in-depth analysis of recommendation outcomes considering fairness aspects. With (nested) filters on user or item attributes, metrics can easily be compared across user and item subgroups. Further, (nested) filters can be used on the dataset level; this way, recommendation outcomes can be compared across several sub-datasets to analyze for differences considering fairness aspects. The software currently features five datasets, 11 metrics, and 21 recommendation algorithms to be used in computational experimentation. It is open source and developed in a modular manner to facilitate extension. The analysis software consists of two components: A software package (FairRecKitLib) for running recommendation algorithms on the available datasets and a web-based user interface (FairRecKitApp) to start experiments, retrieve results of previous experiments, and analyze details. The application also comes with extensive documentation and options for result customization, which makes for a flexible tool that supports in-depth analysis.

KW - FairRecKit

KW - resource

KW - toolkit

KW - analysis software

KW - web-based evaluation

KW - analysis

KW - recommender systems

KW - music

KW - movies

KW - evaluation

KW - web-based

UR - http://www.scopus.com/inward/record.url?scp=85151433181&partnerID=8YFLogxK

U2 - 10.1145/3576840.3578274

DO - 10.1145/3576840.3578274

M3 - Paper

SP - 438

EP - 443

Y2 - 19 March 2023 through 23 March 2023

ER -

FairRecKit: A Web-based analysis software for recommender evaluations

Abstract

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this