Tabular Data Visualization and Machine Learning based Classification for Registered Korean Fishing Vessels

Abstract

Although the demand for analyzing fishing vessel design trends is increasing, collecting relevant datasets remains challenging due to the lack of digitization in most cases. In this paper, we present an analysis of a dataset comprising specifications of over 4,000 registered Korean fishing vessels. The dataset, manually collected and structured in a tabular format, is organized by rows and columns to represent various vessel attributes. We preprocess the data based on the characteristics of each column and apply exploratory data analysis (EDA) techniques to uncover meaningful patterns, correlations, and insights. Statistical correlation analysis is conducted to examine relationships between key features such as vessel dimensions and speed. We also analyze the distribution of multiple variables across different sectors of the fishing industry. Based on the identified feature patterns, we develop machine learning (ML) techniques to classify vessel types—a critical factor in the vessel design process. Specifically, we implement and compare the performance of a decision tree model and a multi-layer perceptron (MLP) model. The results provide a data-driven approach to support decision-making in fishing vessel design and demonstrate a generalized methodology applicable to similar tabular datasets.