Keywords
How to Cite
Abstract
Ushbu maqolada bojxona nazoratida xavf tovarlarni targetlashda sunʻiy intellekt modellarining qoʻllanilishi tadqiq qilingan. DATE (Dual Attentive Tree-aware Embedding) va gATE (boosted Active Trade Exploration) modellari boshqa mashinaviy oʻrganish algoritmlari bilan empirik solishtirma tahlil oʻtkazilgan. Nigeriya bojxonasining besh yillik import maʻlumotlari asosida modellar samaradorligi baholangan. Oʻzbekistonda xavfni boshqarish tizimi (XBT) tajribasi tahlil qilingan va sunʻiy intellekt modellarini amaliyotga joriy etish boʻyicha takliflar ishlab chiqilgan. DATE modeli 92,66% precision koʻrsatkichiga erishgan, gibrid tanlash strategiyalari concept drift muammosini samarali hal qilishi isbotlangan.
References
1. UNCTAD. (2024). Global Trade Update (December 2024). United Nations Conference on Trade and Development. https://unctad.org/publication/global-trade-update-december-2024
2. World Trade Organization. (2023). Trade Facilitation Agreement Database, Article 7.4: Risk Management. https://tfadatabase.org/en/tfa-text/article/7
3.WCO. (2025). Smart Customs Project: Detailed Report on the Adoption of AI/ML in Customs. https://www.wcoomd.org/-/media/wco/public/global/pdf/topics/facilitation/activities-and-programmes/smart-customs/public-version_detailed-report-on-the-adoption-of-ai-and-ml-in-customs.pdf
4. O'zbekiston Respublikasi Prezidentining 2024-yil 14-oktabrdagi PQ-358-son "Sun'iy intellekt texnologiyalarini 2030-yilga qadar rivojlantirish strategiyasini tasdiqlash to'g'risida"gi qarori. https://lex.uz/docs/-7158604
5 O'zbekiston Respublikasi Prezidentining 2025-yil 25-martdagi PF-57-son "Davlat bojxona xizmati organlarining faoliyati samaradorligini oshirish bo'yicha chora-tadbirlar to'g'risida"gi Farmoni. https://lex.uz/uz/docs/-7451997
6. O'zbekiston Respublikasi Prezidentining 2025-yil 25-martdagi PQ-122-son "Davlat bojxona xizmati organlarining faoliyatini tashkil etish va bojxona ma'muriyatchiligini yanada takomillashtirishning qo'shimcha chora-tadbirlari to'g'risida"gi qarori. https://lex.uz/uz/docs/-7452043
7. Kim, S., Tsai, Y-C., Singh, K., Choi, Y., Ibok, E., Li, C-T., & Cha, M. (2020). DATE: Dual Attentive Tree-aware Embedding for Customs Fraud Detection. KDD '20. https://doi.org/10.1145/3394486.3403339
8. Squirrell, M. (2020). Machine Learning for Customs Valuation. WCO BACUDA Technical Report. https://github.com/YSCHOI-github/Customs_Fraud_Detection
9. Vanhoeyveld, J., Martens, D., & Peeters, B. (2020). Customs fraud detection: Assessing the value of behavioural and high-cardinality data under the imbalanced learning issue. Pattern Analysis and Applications. https://doi.org/10.1007/s10044-019-00852-w
10. De Roux, D., Perez, B., Moreno, A., Villamil, M.P., & Figueroa, C. (2018). Tax fraud detection for under-reporting declarations using an unsupervised machine learning approach. KDD '18. https://doi.org/10.1145/3219819.3219878
11.Canrakerta, A., Hidayanto, A.N., & Ruldeviyani, Y. (2020). Application of business intelligence for customs declaration: A case study in Indonesia. Journal of Physics: Conference Series, 1444, 012028. https://doi.org/10.1088/1742-6596/1444/1/012028
12.Kim, S., Mai, T-D., Han, S., Park, S., Nguyen, T.N.D.K., So, J., Singh, K., & Cha, M. (2022). Active Learning for Human-in-the-Loop Customs Inspection. IEEE TKDE. https://doi.org/10.1109/TKDE.2022.3144299
13.Kim, S., Mai, T-D., et al. (2023). gATE: boosted Active Trade Exploration. arXiv:2010.14282. https://arxiv.org/abs/2010.14282
14. WCO & WTO. (2022). Study Report on Disruptive Technologies. https://www.wto.org/english/res_e/booksp_e/wco-wto_e.pdf
15. O'zbekiston Respublikasi Bojxona kodeksi. (2016). 29-bob, 203–207-moddalar. https://lex.uz/docs/-3521436
16. World Customs Organization. (2020). SAFE Framework of Standards. https://www.wcoomd.org/en/topics/facilitation/instrument-and-tools/frameworks-of-standards/safe_package.aspx