The documentary sources for the political, economic, and social history of ancient Mesopotamia constitute hundreds of thousands of clay tablets inscribed in the cuneiform script. Most tablets are damaged, leaving gaps in the texts written on them. Atrahasis is a machine learning model trained on available digitised daily economic and administrative records from Babylonia under the Persian empire (6th-4th cent. BCE). Version 1.0 on the Babylonian Engine Website is based on data prepared by F. Joannès and his team in the framework of the Achemenet program (CNRS, Nanterre); see more on the Achemenet website. Future versions of the model will be trained on more text corpora, and will allow for specific registered users to upload projects and train them on a specific group of texts. Atrahasis can help restore damaged parts of Akkadian transliterations denoted by (currently up to three at a time).
For example: NAME a šú šá NAME ana NAME lú ébabbarra u NAME lú sanga LOCATION iqbi umma
The results rank best suggested restoration. At the moment texts can only entered in the suggested machine readable format, we are working on a more user friendly version - stay tuned. As the training corpus grows, Atrahasis can help restore more text genres and larger sections of broken text. This is a first step towards large-scale reconstruction of the lost ancient Babylonia heritage.
This repository is made freely available under the Attribution-ShareAlike 3.0 Unported (CC BY-SA 3.0) license. This means you are free to share and adapt the code and datasets, under the conditions that you cite the project appropriately, note any changes you have made to the original code and datasets, and if you are redistributing the project or a part thereof, you must release it under the same license or a similar one.
For more information about the license, see here.
If you are experiencing any issues with the website or the git repository, please contact us at dhl.arieluni@gmail.com, and we would gladly assist you. We would also much appreciate feedback about using the code via the website, or about the repository itself, so please send us any comments or suggestions.
-
Shai Gordin
-
Ethan Fetaya
This research was supported by the Ministry of Science & Technology ,Israel.