Model Prediksi Umur Kepiting Berdasarkan Data Morfometrik dan Gender: Pendekatan Model Support Vector Regression

Abstract

Crab is one of the most important marine commodities and resources in Indonesian waters, both economically and ecologically. Crab age determination can provide a better understanding of crab growth and development, so that crab farming can be carried out effectively, efficiently, and profitably for its supporters. In addition, determining the age of the crab can also help the sustainability of the crab population. This study was conducted using support vector regression (SVR) modeling to predict crab age by establishing a predictive relationship between the dependent variable (x) and the independent variable (y). The attributes of the dependent variable (x) include length, diameter, height, and weight. While the independent variable (y) only includes crab age. SVR modeling is carried out to show predicted data with actual data, where the results of the SVR modeling will be evaluated based on the results of the RMSE value test. This study resulted in an RMSE value of 0.019814 so it can be said that the model to predict crab age is very accurate. The purpose of this research is to develop a statistical model that can predict crab age based on morphometric data and crab gender using the Support Vector Regression (SVR) model approach.