Article Per Year (5 Year)
p-Index From 2021 - 2026
0.23
P-Index
This Author published in this journals
All Journal Jurnal Pengelolaan Perikanan Tropis (Journal Of Tropical Fisheries Management)
citra amelia nawati
Universitas Pendidikan Indonesia
Agriculture, Biological Sciences & Forestry Biochemistry, Genetics & Molecular Biology Decision Sciences, Operations Research & Management Earth & Planetary Sciences Environmental Science

Published : 1 Documents Claim Missing Document
Claim Missing Document
Check
Articles

Found 1 Documents
Search

 1 
Performance Analysis of the K-Nearest Neighbors s (KNN) Algorithm on Classification of Small Pelagic Fish Abundance in the East Season in Banten Waters citra amelia nawati; Ayang Armelita Rosalia; Novi Sofia Fitriasari
Jurnal Pengelolaan Perikanan Tropis (Journal of Tropical Fisheries Management) Vol 10 No 1 (2026): Jurnal Pengelolaan Perikanan Tropis (Journal Of Tropical Fisheries Management)
Publisher : Departement of Aquatic Resources Management, Faculty of Fisheries and Marine Sciences, IPB University

Show Abstract | Download Original | Original Source | Check in Google Scholar | DOI: 10.29244/jppt.10.1.24-35

Abstract

Small pelagic fish are the dominant commodity in the fishing sector in the waters of Banten, accounting for 51.3% of total production in 2023. However, limited access to technology means that conventional fishermen in this region still rely on experience and intuition to determine fishing locations, which results in unstable catch yields and inefficiencies in operational costs, time, and labor. This study is designed to analyze the performance of the K-Nearest Neighbors (KNN) algorithm in classifying the abundance of small pelagic fish during the eastern season in the waters of Banten. Two oceanographic parameters were used as predictors: Sea Surface Temperature (SST) and chlorophyll-a concentration, both of which serve as indicators of aquatic ecosystem productivity. Secondary data were sourced from Aqua MODIS satellite imagery spanning the years 2021–2025, which were subsequently processed and integrated via the Kaggle and Google Colab platforms. The final dataset comprises 3,382 data points divided into two classes: potential and non-potential. Model testing was conducted by varying the K value (3, 5, 7, and 9) and the data split ratio (60:40, 70:30, and 80:20), using Euclidean Distance and the Confusion Matrix as performance evaluation metrics. The model with a configuration of K=7 and a 70:30 split ratio proved to yield the best performance, characterized by an accuracy of 95.66% and proportional precision, recall, and F1-score values across both classes. Permutation Feature Importance analysis indicates that chlorophyll-a contributes dominantly at 25.66%, while the recorded influence of SPL is 23.58%. These results confirm the superiority of the KNN algorithm in classifying small pelagic fish in the waters of Banten.
Co-Authors Ayang Armelita Rosalia Novi Sofia Fitriasari