Explainable Product Classification for Customs
Not on the map. Why?
Abstract (arXiv)
The task of assigning internationally accepted commodity codes (aka HS codes) to traded goods is a critical function of customs offices. Like court decisions made by judges, this task follows the doctrine of precedent and can be nontrivial even for experienced officers. Together with the Korea Customs Service (KCS), we propose a first-ever explainable decision supporting model that suggests the most likely subheadings (i.e., the first six digits) of the HS code. The model also provides reasoning for its suggestion in the form of a document that is interpretable by customs officers. We evaluated the model using 5,000 cases that recently received a classification request. The results showed that the top-3 suggestions made by our model had an accuracy of 93.9% when classifying 925 challenging subheadings. A user study with 32 customs experts further confirmed that our algorithmic suggestions accompanied by explainable reasonings, can substantially reduce the time and effort taken by customs officers for classification reviews.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 22 of its 51 references that are in the map. First posted on arXiv 18 Nov 2023.
Builds on
22 of 51 references.
References · 22
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2017On Calibration of Modern Neural Networkscited by 4,860
- 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
- 2016Bidirectional Attention Flow for Machine Comprehensioncited by 278
- 2019QASC: A Dataset for Question Answering via Sentence Compositioncited by 248
- 2021What Do We Want From Explainable Artificial Intelligence (XAI)? -- A Stakeholder Perspective on XAI and a Conceptual Model Guiding Interdisciplinary XAI Researchcited by 119
- 2021KLUE: Korean Language Understanding Evaluationcited by 113
- 2019Hierarchical Graph Network for Multi-hop Question Answeringcited by 83
- 2019Hyperbolic Interaction Model For Hierarchical Multi-Label Classificationcited by 30
- 2020A Simple Yet Strong Pipeline for HotpotQAcited by 24
- 2020Unsupervised Alignment-based Iterative Evidence Retrieval for Multi-hop Question Answeringcited by 24
- 2022Deconfounding Legal Judgment Prediction for European Court of Human Rights Cases Towards Better Alignment with Expertscited by 24
10 more
- 2021MATCH: Metadata-Aware Text Classification in A Large Hierarchycited by 20
- 2019Evidence Sentence Extraction for Machine Reading Comprehensioncited by 14
- 2019Identifying Supporting Facts for Multi-hop Question Answering with Document Graph Networkscited by 7
- 2021Self-interpretable Convolutional Neural Networks for Text Classificationcited by 5
- 2020iCapsNets: Towards Interpretable Capsule Networks for Text Classificationcited by 4
- 2020Active Learning for Human-in-the-Loop Customs Inspectioncited by 4
- 2021Customs Fraud Detection in the Presence of Concept Driftcited by 4
- 2016Bridging the Gap: Incorporating a Semantic Similarity Measure for Effectively Mapping PubMed Queries to Documentscited by 3
- 2021Classification of Goods Using Text Descriptions With Sentences Retrievalcited by 3
- 2019Interpretable Text Classification Using CNN and Max-poolingcited by 1
Built on by
Citing papers per year: 2026: 2.
2 papers.