Please use this identifier to cite or link to this item:
|Title:||Exploring Machine Learning Models to Predict Harmonized System Code|
|Other Titles:||استكشاف نماذج التعلم الآلي للتنبؤ بكود النظام المنسق|
|Authors:||AL Taheri, Fatma|
United Arab Emirates (UAE)
|Publisher:||The British University in Dubai (BUiD)|
|Abstract:||Globalization has shaped the way governments and government agencies operate; alongside how said phenomenon has consequently paved the way toward economic growth. With globalization, use of modern technology has also become a vital component in the public sector. Customs, for one, recognizes the importance of technology in ensuring efficiency of international trade. The Harmonized System (HS) Code is widely used across all customs departments because of the several benefits it yields for the government agency including a more convenient and easier approach for calculating fees and taxes. In that regard, it is the purpose of this study to explore ways to reduce the complexity, gaps and many other challenges in using HS Code in Dubai Customs, UAE using a case study approach and a machine learning-based HS Code Prediction model. This study uses six machine learning models based on the CRISP-DM framework. Initially, the study acquires the datasets from Dubai Customs and then analyses the data. Following this is the preparation of data for processing and the creation of the machine learning models. The results of the study indicate that machine learning models are effective tools in predicting HS Code for the user goods descriptions. In this study, six machine learning-based models have been implemented to determine the ability of detecting the HS Code based on the user’s input description, where the highest achieved accuracy is 76.3% using linear support vector machine model.|
|Appears in Collections:||Dissertations for Informatics (Knowledge and Data Management)|
Files in This Item:
There are no files associated with this item.
Items in DSpace are protected by copyright, with all rights reserved, unless otherwise indicated.