MODELLING SARDINE FISHING IN PAPUA USING COMMON NEAREST NEIGHBOR CLUSTERING

Abstrak

Papua and Papua Barat provinces, Indonesia, possess a rich marine tapestry woven with the thread of sardine fish, a cornerstone of commercial fisheries. Understanding the spatial distribution of sardine catches is crucial for sustainable resource management and economic development. This study investigates the application of Common Nearest Neighbor Clustering (CNNC) on Sardines catch data from Papua and Papua Barat Province. To address potential multi-collinearity among these attributes, Principal Component Analysis (PCA) was employed as a preprocessing step. The clustering algorithm was optimized with an epsilon parameter of 0.65 and a leaf size of 30, yielding a silhouette score of 0.224373, which indicates moderate clustering quality. The analysis resulted in the identification of seven distinct clusters within the data, providing valuable insights into the distribution and characteristics of Sardines catches across the region. The findings contribute to the understanding of fisheries management in Papua and Papua Barat, with implications for policy and resource allocation. The results of this research contribute to a nuanced understanding of sardine catch distribution in Papua and Papua Barat. By identifying regional clusters, policymakers can tailor fisheries management strategies to specific needs, ensuring the long-term sustainability of this vital resource. Furthermore, the findings provide insights for stakeholders in the fishing industry to optimize operations and enhance economic benefits. This study underscores the importance of spatial analysis in unraveling the complexities of marine ecosystems and supports informed decision-making for the sustainable utilization of marine resources.