Application of the K-Nearest Neighbor algorithm as a determination of the quality of ornamental fishing gear.

Autor: Kasmi, Mauli, Aman, Andryanto, Mariam, M., Iskandar, Akbar, Sonodihardjo, Asti Sugiarti, Parewe, Andi Maulidinnawati Abdul Kadir, Sulkifli, S., Hadi, Syamsul
Předmět:
Zdroj: AIP Conference Proceedings; 2024, Vol. 3098 Issue 1, p1-11, 11p
Abstrakt: This study aims to determine the level of quality of ornamental fish catches from fishermen based on the criteria for the friendliness of fishing gear used by fishermen. This study used the Rank Order Centroid (ROC) method with a weighting criterion value of 0.52 = Good, 0.27 = Moderate, 0.15 = Low, and 0.06 = Very Low. Then classification is carried out using the K-Nearest Neighbor method as a determination of the quality of ornamental fish catches from fishermen. The results showed that from 20 training data which were fishermen who used ornamental fish catches which were then tested with 5 new fishermen data from ornamental fish catches, obtained accuracy or suitability of data both received and rejected from ornamental fish collection companies. This proves that the results of the Centroid Rank Order method and the K-Nearest Neighbor method are much better in determining the quality of ornamental fishing gear used by fishermen. [ABSTRACT FROM AUTHOR]
Databáze: Complementary Index