EUVIMEDEUVIMED entdecken
Open Access · CC BY-SA 4.0

Analysis of the K-Nearest Neighbor (K-NN) Algorithm Method for Student Extracurricular Recommendations

A. Nurul Anwar, Dani Dani, Chandra Alim

bit-Tech · 10.08.2026 · 9 (1) · 1567-1577

DOI: 10.32877/bt.v9i1.4146 ↗

Kurz gefasst

Fachartikel in der Zeitschrift bit-Tech, online erschienen am 10.08.2026. Verfasst von A. Nurul Anwar und 2 weiteren Autor:innen. Der Volltext ist frei zugänglich (Open Access, CC BY-SA 4.0).

AbstractBibliografische AngabenZitierenVerwandte Studien

Abstract (Originalsprache)

This study examined whether the K-Nearest Neighbor (K-NN) algorithm could reconstruct historical assignments of students to three sports extracurricular activities—futsal, badminton, and volleyball—rather than independently determine extracurricular suitability. The analysis used 70 records from Pesantren Nafidatunnajah, comprising 24 futsal, 24 badminton, and 22 volleyball cases. A fixed partition of 50 training records (17 futsal, 17 badminton, and 16 volleyball) and 20 testing records (7 futsal, 7 badminton, and 6 volleyball) was used; however, no random or stratified selection procedure was documented. Five predictors were included: sports score, height, agility, stamina, and sports interest. The variables were numerically encoded and processed for Euclidean-distance classification, although the nominal sports-interest codes (futsal = 1, badminton = 2, volleyball = 3) directly corresponded to the target categories and may therefore function as a target proxy. Using K = 5 and majority voting, the model correctly classified 18 of 20 test records, yielding 90% accuracy. Because the test set contained only 20 observations, each error changed accuracy by five percentage points, and the corresponding 95% Wilson confidence interval was approximately 69.9%–97.2%. Accordingly, the result should be interpreted as preliminary evidence that K-NN can reproduce historical extracurricular labels within this small, single-school dataset, not as stable evidence that the model identified an integrated relationship among students’ physical characteristics, performance, and preferences. Validation without the sports-interest variable, ablation analysis, repeated resampling, and larger multisite datasets are required before the model can support defensible extracurricular placement decisions.

Zitieren

APA

Anwar, A. N., Dani, D., & Alim, C. (2026). Analysis of the K-Nearest Neighbor (K-NN) Algorithm Method for Student Extracurricular Recommendations. bit-Tech, 9(1), 1567-1577. https://doi.org/10.32877/bt.v9i1.4146

Vancouver

Anwar AN, Dani D, Alim C. Analysis of the K-Nearest Neighbor (K-NN) Algorithm Method for Student Extracurricular Recommendations. bit-Tech. 2026;9(1):1567-1577. doi:10.32877/bt.v9i1.4146

BibTeX
@article{Anwar2026,
  author = {Anwar, A. Nurul and Dani, Dani and Alim, Chandra},
  title = {Analysis of the K-Nearest Neighbor (K-NN) Algorithm Method for Student Extracurricular Recommendations},
  journal = {bit-Tech},
  year = {2026},
  volume = {9},
  number = {1},
  pages = {1567--1577},
  doi = {10.32877/bt.v9i1.4146}
}
Kostenlos starten →