A Fairness-Aware Semi-Supervised Clustering Method

Cristina Maier; Cyrus Saadat; Dan Simovici

doi:10.32473/flairs.39.1.141842

A Fairness-Aware Semi-Supervised Clustering Method

Cristina Maier
, Cyrus Saadat
, Dan Simovici

Computer Science

Boston College
University of Massachusetts Boston

Research output: Contribution to journal › Conference article › peer-review

Abstract

We present a semi-supervised clustering algorithm that incorporates a fairness component, implemented as a variant of K-Means but extendable to other center-based approaches. Fairness is defined as producing balanced clusters and is measured using a normalized entropy metric. Experiments on real-world and LLM-generated datasets show consistent improvements in fairness and accuracy over baseline K-Means, along with an analysis of the effect of the fairness component strength.

Original language	English
Journal	Proceedings of the International Florida Artificial Intelligence Research Society Conference, FLAIRS
Volume	39
Issue number	1
DOIs	https://doi.org/10.32473/flairs.39.1.141842
State	Published - 2026
Event	39th International Florida Artificial Intelligence Research Society Conference, FLAIRS-39 2026 - Marco Island, United States Duration: May 17 2026 → May 20 2026

ASJC Scopus Subject Areas

Software
Artificial Intelligence

Access to Document

10.32473/flairs.39.1.141842

Cite this

@article{1a39fd7b11b346e3a4984885f45603ec,

title = "A Fairness-Aware Semi-Supervised Clustering Method",

abstract = "We present a semi-supervised clustering algorithm that incorporates a fairness component, implemented as a variant of K-Means but extendable to other center-based approaches. Fairness is defined as producing balanced clusters and is measured using a normalized entropy metric. Experiments on real-world and LLM-generated datasets show consistent improvements in fairness and accuracy over baseline K-Means, along with an analysis of the effect of the fairness component strength.",

author = "Cristina Maier and Cyrus Saadat and Dan Simovici",

note = "Publisher Copyright: {\textcopyright} 2026, Florida Online Journals, University of Florida. All rights reserved.; 39th International Florida Artificial Intelligence Research Society Conference, FLAIRS-39 2026 ; Conference date: 17-05-2026 Through 20-05-2026",

year = "2026",

doi = "10.32473/flairs.39.1.141842",

language = "English",

volume = "39",

journal = "Proceedings of the International Florida Artificial Intelligence Research Society Conference, FLAIRS",

issn = "2334-0754",

publisher = "Florida Online Journals, University of Florida",

number = "1",

}

TY - JOUR

T1 - A Fairness-Aware Semi-Supervised Clustering Method

AU - Maier, Cristina

AU - Saadat, Cyrus

AU - Simovici, Dan

PY - 2026

Y1 - 2026

N2 - We present a semi-supervised clustering algorithm that incorporates a fairness component, implemented as a variant of K-Means but extendable to other center-based approaches. Fairness is defined as producing balanced clusters and is measured using a normalized entropy metric. Experiments on real-world and LLM-generated datasets show consistent improvements in fairness and accuracy over baseline K-Means, along with an analysis of the effect of the fairness component strength.

AB - We present a semi-supervised clustering algorithm that incorporates a fairness component, implemented as a variant of K-Means but extendable to other center-based approaches. Fairness is defined as producing balanced clusters and is measured using a normalized entropy metric. Experiments on real-world and LLM-generated datasets show consistent improvements in fairness and accuracy over baseline K-Means, along with an analysis of the effect of the fairness component strength.

UR - https://www.scopus.com/pages/publications/105040191022

UR - https://www.scopus.com/pages/publications/105040191022#tab=citedBy

U2 - 10.32473/flairs.39.1.141842

DO - 10.32473/flairs.39.1.141842

M3 - Conference article

AN - SCOPUS:105040191022

SN - 2334-0754

VL - 39

JO - Proceedings of the International Florida Artificial Intelligence Research Society Conference, FLAIRS

JF - Proceedings of the International Florida Artificial Intelligence Research Society Conference, FLAIRS

IS - 1

T2 - 39th International Florida Artificial Intelligence Research Society Conference, FLAIRS-39 2026

Y2 - 17 May 2026 through 20 May 2026

ER -

A Fairness-Aware Semi-Supervised Clustering Method

Abstract

ASJC Scopus Subject Areas

Access to Document

Other files and links

Fingerprint

Cite this