PENERAPAN METODE REGRESI DALAM ANALISIS TINGKAT KONSUMSI IKAN DI JAWA TIMUR

The Indonesian marine ecosystem, rich in water resources, makes the fishing sector important for the economy, food, and jobs. East Java, with the highest catch production in Indonesia, has huge potential in the fishing sector. Despite its high potential, fish consumption in Indonesia is still relatively low due to lack of awareness, unoptimal distribution, and other factors. The study aims to analyze the level of fish consumption in Eastern Java and determine the best regression method to predict the rate of fish intake based on the region and the types of available commodities. This study uses three regression methods, namely Linear Regression, Support Vector Regression (SVR), and Gradient Boosting Machine. (GBM). Data visualization is done using bar diagrams, and results are validated using the determination coefficient R2 which is then analyzed descriptively. The Sumenep region has the highest level of fish consumption in East Java during the period 2018-2020. Whereas the commodities with the highest consumption are Tuna, Tongkol, Cakalang (TTC) Diawetkan. The GBM method showed its best performance and proved to be the most effective and accurate in predicting the level of fish consumption in East Java with a perfect determination coefficient (0,9999), compared to Linear Regression (0,8755) and SVR (0,9825).