Predictive Comparative Analysis for Fundamental Risk of U.S Import and Economy Seafood Market

Future obstacles include the growth of illegal, unreported, and unregulated (IUU) crimes and the presence of a market leader in the US as the main fish importer. Almost of the world's marine fish stocks have already reached total depletion and are in danger of going extinct, according to the United Nations Conference on Trade and Development (UNCTAD, 2018). Fishing that is illegal, unreported, and unregulated (IUU) is a serious issue here. The facts indicate that exporting countries engage in (IUU) practices by processing seafood from numerous different fleets. Therefore, smart technology based on innovative data and analytics systems such as Machine Learning technology is needed to enable fully monitored fisheries. This research aims to provide the best analysis method, focusing on the fundamental risk characteristics of marine capture. The research objective is to analyze the performance of the best model and predict high-risk fundamental characteristics in countries exporting to the U.S. The data utilized consists of 115 samples obtained from the United States International Trade Commission (USITC). The results of the method analysis, various algorithms, including K-Nearest Neighbors (KNN), Logistic Regression, and Decision Tree, were considered for comparison. However, the results indicate that Random Forest stands out as the most effective method, demonstrating the highest predictive capability for high-level fundamental risk. This is evident from the impressive AUC value of 0.993, an Accuracy score of 0.9222, Precision of 0.9510, and Recall value of 0.9571.