To run this code with your data, please make your preprocessor and modify data loader.
| Data | # Imports | Illicit rate | Labeled ratio | Period | Misc |
|---|---|---|---|---|---|
| synthetic | 0.1M | 7.58% | 100% | Jan 13 – Dec 13 | With the repo |
| synthetic-k | 0.1M | 23.0% | 100% | Jan 20 - Dec 20 | Confidential |
| synthetic-k-partial | 0.1M | 23.0% | 10% | Jan 20 - Dec 20 | Confidential |
| real-m | 0.42M | 1.64% | 100% | Jan 13 – Dec 16 | |
| real-c | 1.90M | 1.21% | 100% | Jan 16 – Dec 19 | |
| real-n | 1.93M | 4.12% | 100% | Jan 13 – Dec 17 | |
| real-t | 4.17M | 8.16% | 100% | Jan 15 – Dec 19 |
To run this code with the refined synthetic datasets (synthetic-k, synthetic-k-partial), please contact Sundong Kim (sundong@gist.ac.kr) with the Confidentiality Statement. If you want to work on real datasets (real-n, real-m, real-t, real-c) and please directly consult with the data providers (Sungsig Kim, sungsig.kim@wcoomd.org, World Customs Organization)