Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm

Aliya Kalykulova; Bilal Saoud; Ibraheem Shayea; Dauren Sagidullauly

doi:10.1007/978-3-032-00232-7_24

Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm

Aliya Kalykulova
, Bilal Saoud^*
, Ibraheem Shayea
, Dauren Sagidullauly

^*Corresponding author for this work

Faculty of Electrical and Electronics

Astana IT University
Akli Mohand Oulhadj University of Bouira

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

Abstract

Effective extraction of domain-specific terms and named entities is a key challenge in text mining. This paper investigates the use of the k-means clustering algorithm for unsupervised extraction of unigrams and named entities from text data. The approach groups terms based on their vector representations, enabling the identification of semantically similar words without labeled data. Experiments conducted on the ACTER (Annotated Corpora for Term Extraction Research) corpus evaluate the method using precision, recall, and F1-score. Results show average scores of 25.79% precision, 40.05% recall and 30.47% F1-score, with optimal performance achieved using 40 to 60 clusters. Future work will explore algorithm optimization and comparisons with alternative extraction techniques.

Original language	English
Title of host publication	Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations
Editors	Ali Othman Albaji
Publisher	Springer Science and Business Media Deutschland GmbH
Pages	375-386
Number of pages	12
ISBN (Print)	9783032002310
DOIs	https://doi.org/10.1007/978-3-032-00232-7_24
Publication status	Published - 2026
Event	International Conference on AI: Current Research, Industry Trends, and Innovations, FICAILY 2025 - Tripoli, Libya Duration: 9 Jul 2025 → 10 Jul 2025

Publication series

Name	Studies in Computational Intelligence
Volume	1229 SCI
ISSN (Print)	1860-949X
ISSN (Electronic)	1860-9503

Conference

Conference	International Conference on AI: Current Research, Industry Trends, and Innovations, FICAILY 2025
Country/Territory	Libya
City	Tripoli
Period	9/07/25 → 10/07/25

Bibliographical note

Publisher Copyright:
© The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

Keywords

Automated term extraction
Clustering
NLP
Unigrams
Unsupervised annotator
k-means

Access to Document

10.1007/978-3-032-00232-7_24

Cite this

Kalykulova, A., Saoud, B., Shayea, I., & Sagidullauly, D. (2026). Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm. In A. O. Albaji (Ed.), Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations (pp. 375-386). (Studies in Computational Intelligence; Vol. 1229 SCI). Springer Science and Business Media Deutschland GmbH. https://doi.org/10.1007/978-3-032-00232-7_24

Kalykulova, Aliya ; Saoud, Bilal ; Shayea, Ibraheem et al. / Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm. Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations. editor / Ali Othman Albaji. Springer Science and Business Media Deutschland GmbH, 2026. pp. 375-386 (Studies in Computational Intelligence).

@inproceedings{060c98b61f934342b3cb65a650e0e563,

title = "Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm",

abstract = "Effective extraction of domain-specific terms and named entities is a key challenge in text mining. This paper investigates the use of the k-means clustering algorithm for unsupervised extraction of unigrams and named entities from text data. The approach groups terms based on their vector representations, enabling the identification of semantically similar words without labeled data. Experiments conducted on the ACTER (Annotated Corpora for Term Extraction Research) corpus evaluate the method using precision, recall, and F1-score. Results show average scores of 25.79\% precision, 40.05\% recall and 30.47\% F1-score, with optimal performance achieved using 40 to 60 clusters. Future work will explore algorithm optimization and comparisons with alternative extraction techniques.",

keywords = "Automated term extraction, Clustering, NLP, Unigrams, Unsupervised annotator, k-means",

author = "Aliya Kalykulova and Bilal Saoud and Ibraheem Shayea and Dauren Sagidullauly",

note = "Publisher Copyright: {\textcopyright} The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.; International Conference on AI: Current Research, Industry Trends, and Innovations, FICAILY 2025 ; Conference date: 09-07-2025 Through 10-07-2025",

year = "2026",

doi = "10.1007/978-3-032-00232-7\_24",

language = "English",

isbn = "9783032002310",

series = "Studies in Computational Intelligence",

publisher = "Springer Science and Business Media Deutschland GmbH",

pages = "375--386",

editor = "Albaji, \{Ali Othman\}",

booktitle = "Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations",

address = "Germany",

}

Kalykulova, A, Saoud, B, Shayea, I & Sagidullauly, D 2026, Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm. in AO Albaji (ed.), Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations. Studies in Computational Intelligence, vol. 1229 SCI, Springer Science and Business Media Deutschland GmbH, pp. 375-386, International Conference on AI: Current Research, Industry Trends, and Innovations, FICAILY 2025, Tripoli, Libya, 9/07/25. https://doi.org/10.1007/978-3-032-00232-7_24

Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm. / Kalykulova, Aliya; Saoud, Bilal; Shayea, Ibraheem et al.
Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations. ed. / Ali Othman Albaji. Springer Science and Business Media Deutschland GmbH, 2026. p. 375-386 (Studies in Computational Intelligence; Vol. 1229 SCI).

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

TY - GEN

T1 - Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm

AU - Kalykulova, Aliya

AU - Saoud, Bilal

AU - Shayea, Ibraheem

AU - Sagidullauly, Dauren

N1 - Publisher Copyright: © The Author(s), under exclusive license to Springer Nature Switzerland AG 2026.

PY - 2026

Y1 - 2026

N2 - Effective extraction of domain-specific terms and named entities is a key challenge in text mining. This paper investigates the use of the k-means clustering algorithm for unsupervised extraction of unigrams and named entities from text data. The approach groups terms based on their vector representations, enabling the identification of semantically similar words without labeled data. Experiments conducted on the ACTER (Annotated Corpora for Term Extraction Research) corpus evaluate the method using precision, recall, and F1-score. Results show average scores of 25.79% precision, 40.05% recall and 30.47% F1-score, with optimal performance achieved using 40 to 60 clusters. Future work will explore algorithm optimization and comparisons with alternative extraction techniques.

AB - Effective extraction of domain-specific terms and named entities is a key challenge in text mining. This paper investigates the use of the k-means clustering algorithm for unsupervised extraction of unigrams and named entities from text data. The approach groups terms based on their vector representations, enabling the identification of semantically similar words without labeled data. Experiments conducted on the ACTER (Annotated Corpora for Term Extraction Research) corpus evaluate the method using precision, recall, and F1-score. Results show average scores of 25.79% precision, 40.05% recall and 30.47% F1-score, with optimal performance achieved using 40 to 60 clusters. Future work will explore algorithm optimization and comparisons with alternative extraction techniques.

KW - Automated term extraction

KW - Clustering

KW - NLP

KW - Unigrams

KW - Unsupervised annotator

KW - k-means

UR - https://www.scopus.com/pages/publications/105028287079

U2 - 10.1007/978-3-032-00232-7_24

DO - 10.1007/978-3-032-00232-7_24

M3 - Conference contribution

AN - SCOPUS:105028287079

SN - 9783032002310

T3 - Studies in Computational Intelligence

SP - 375

EP - 386

BT - Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations

A2 - Albaji, Ali Othman

PB - Springer Science and Business Media Deutschland GmbH

T2 - International Conference on AI: Current Research, Industry Trends, and Innovations, FICAILY 2025

Y2 - 9 July 2025 through 10 July 2025

ER -

Kalykulova A, Saoud B, Shayea I, Sagidullauly D. Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm. In Albaji AO, editor, Selected Papers from the International Conference on Artificial Intelligence - FICAILY2025 - Current Research, Industry Trends, and Innovations. Springer Science and Business Media Deutschland GmbH. 2026. p. 375-386. (Studies in Computational Intelligence). doi: 10.1007/978-3-032-00232-7_24

Automatic Unsupervised Extraction of Unigrams of Terms and Named Entities Using the K-Means Clustering Algorithm

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Other files and links

Fingerprint

Cite this