View metadata

dc.titleAutomatic Product Classification in International Trade: Machine Learning and Large Language Models
dc.contributor.authorMarra de Artiñano, Ignacio
dc.contributor.authorRiottini Depetris, Franco
dc.contributor.authorVolpe Martincus, Christian
dc.contributor.orgunitProductivity, Trade and Innovation Sector
dc.contributor.orgunitTrade and Investment Division
dc.coverageLatin America
dc.date.available2023-07-21T00:07:00
dc.date.issue2023-07-21T00:07:00
dc.description.abstractAccurately classifying products is essential in international trade. Virtually all countries categorize products into tariff lines using the Harmonized System (HS) nomenclature for both statistical and duty collection purposes. In this paper, we apply and assess several different algorithms to automatically classify products based on text descriptions. To do so, we use agricultural product descriptions from several public agencies, including customs authorities and the United States Department of Agriculture (USDA). We find that while traditional machine learning (ML) models tend to perform well within the dataset in which they were trained, their precision drops dramatically when implemented outside of it. In contrast, large language models (LLMs) such as GPT 3.5 show a consistently good performance across all datasets, with accuracy rates ranging between 60% and 90% depending on HS aggregation levels. Our analysis highlights the valuable role that artificial intelligence (AI) can play in facilitating product classification at scale and, more generally, in enhancing the categorization of unstructured data.
dc.format.extent37
dc.identifier.doihttp://dx.doi.org/10.18235/0005012
dc.identifier.urlhttps://publications.iadb.org/publications/english/document/Automatic-Product-Classification-in-International-Trade-Machine-Learning-and-Large-Language-Models.pdf
dc.language.isoen
dc.publisherInter-American Development Bank
dc.subjectExport of Goods
dc.subjectCustoms Administration
dc.subjectInternational Trade
dc.subjectMachine Learning
dc.subjectSmall Business
dc.subjectArtificial Intelligence
dc.subjectIntegration and Trade
dc.subjectRating
dc.subjectTariff System
dc.subject.jelcodeF10 - Trade: General
dc.subject.jelcodeC55 - Large Data Sets: Modeling and Analysis
dc.subject.jelcodeC81 - Methodology for Collecting, Estimating, and Organizing Microeconomic Data • Data Access
dc.subject.jelcodeC88 - Other Computer Software
dc.subject.keywordsProduct Classification;machine learning;Large Language Models;Trade
dc.typeWorking Papers
idb.identifier.pubnumberIDB-WP-01494
idb.operationRG-E1716
Return to Publication