Customs Tariff Classification of Organic Chemicals with Data Mining Techniques

Harmonized System (HS Code) is a widely used classification system for commodities and facilitating international trade with 97 chapters according to commodity types. Classifying some of the 97 chapters of the HS Code can be a time-consuming and demanding process. In particular, the classification of organic chemicals belonging to HS Code Chapter 29 is especially challenging and requires specialized knowledge. This research aims to examine the classification of the first four digits of HS Code Chapter 29 of organic chemicals (headings 29.01-29.35) using IUPAC names, a globally recognized chemical nomenclature and terminology. In this study, two machine learning models, K-Nearest Neighbor (KNN) and Support Vector Machines (SVMs), were employed and their parameters were optimized for maximum accuracy. Results showed that the SVM model with optimized parameters achieved the highest accuracy. The results of this study can be applied to improve the efficiency and accuracy of customs operations in the future.