<![CDATA[The economic condition of traditional fishermen is still in a cycle of poverty, so solutions are needed to improve welfare. One solution is to use information technology regarding fishing ground so that fishermen can save fuel and increase the number of catches. Fishing ground information can be determined by processing satellite image data and using machine learning technology. This research aims to create a model that can classify fishing ground using Random Forest and Support Vector Machine algorithms using satellite image data of the Java Sea and its surroundings from 2019-2021 with the parameters chlorophyll, sea surface temperature, salinity, height of the sea, and water temperature. This research shows that the chlorophyll parameter has the greatest role (77.14%) in determining fishing ground. The precision value produced by the Support Vector Machine algorithm (99.83%) is higher than that produced by the Random Forest algorithm (99.80%). However, the classification model produced by the Random Forest algorithm has higher accuracy (99.90%), recall (100%) and F1 score (99.90%) compared to that produced by the Support Vector Machine algorithm, with an accuracy value of (99.89%), recall (99.96%) and F1 score (99.89%).]]>
Copyrights © 2024
