Restoration of fragmentary Babylonian texts using recurrent neural networks

Ethan Fetaya; Yonatan Lifshitz; Elad Aaron; Shai Gordin

doi:10.1073/pnas.2003794117

Restoration of fragmentary Babylonian texts using recurrent neural networks

Ethan Fetaya, Yonatan Lifshitz, Elad Aaron, Shai Gordin

Department of Land of Israel Studies and Archaeology

Research output: Contribution to journal › Article › peer-review

27 Scopus citations

Abstract

The main sources of information regarding ancient Mesopotamian history and culture are clay cuneiform tablets. Many of these tablets are damaged, leading to missing information. Currently, the missing text is manually reconstructed by experts. We investigate the possibility of assisting scholars, by modeling the language using recurrent neural networks and automatically completing the breaks in ancient Akkadian texts from Achaemenid period Babylonia.

Original language	English
Pages (from-to)	22743-22751
Number of pages	9
Journal	Proceedings of the National Academy of Sciences of the United States of America
Volume	117
Issue number	37
DOIs	https://doi.org/10.1073/pnas.2003794117 https://doi.org/10.1073/pnas.2003794117
State	Published - 15 Sep 2020

Keywords

Babylonian heritage
cuneiform script
Late Babylonian dialect
Achaemenid empire
neural networks

Access to Document

Cite this

@article{38a4fb2c8c57452c9d3ad90dd817ecad,

title = "Restoration of fragmentary Babylonian texts using recurrent neural networks",

abstract = "The main sources of information regarding ancient Mesopotamian history and culture are clay cuneiform tablets. Many of these tablets are damaged, leading to missing information. Currently, the missing text is manually reconstructed by experts. We investigate the possibility of assisting scholars, by modeling the language using recurrent neural networks and automatically completing the breaks in ancient Akkadian texts from Achaemenid period Babylonia.",

keywords = "Babylonian heritage, cuneiform script, Late Babylonian dialect, Achaemenid empire, neural networks",

author = "Ethan Fetaya and Yonatan Lifshitz and Elad Aaron and Shai Gordin",

year = "2020",

month = sep,

day = "15",

doi = "10.1073/pnas.2003794117",

language = "אנגלית",

volume = "117",

pages = "22743--22751",

journal = "Proceedings of the National Academy of Sciences of the United States of America",

issn = "0027-8424",

publisher = "National Academy of Sciences",

number = "37",

}

TY - JOUR

T1 - Restoration of fragmentary Babylonian texts using recurrent neural networks

AU - Fetaya, Ethan

AU - Lifshitz, Yonatan

AU - Aaron, Elad

AU - Gordin, Shai

PY - 2020/9/15

Y1 - 2020/9/15

N2 - The main sources of information regarding ancient Mesopotamian history and culture are clay cuneiform tablets. Many of these tablets are damaged, leading to missing information. Currently, the missing text is manually reconstructed by experts. We investigate the possibility of assisting scholars, by modeling the language using recurrent neural networks and automatically completing the breaks in ancient Akkadian texts from Achaemenid period Babylonia.

AB - The main sources of information regarding ancient Mesopotamian history and culture are clay cuneiform tablets. Many of these tablets are damaged, leading to missing information. Currently, the missing text is manually reconstructed by experts. We investigate the possibility of assisting scholars, by modeling the language using recurrent neural networks and automatically completing the breaks in ancient Akkadian texts from Achaemenid period Babylonia.

KW - Babylonian heritage

KW - cuneiform script

KW - Late Babylonian dialect

KW - Achaemenid empire

KW - neural networks

UR - http://www.scopus.com/inward/record.url?scp=85091072990&partnerID=8YFLogxK

U2 - 10.1073/pnas.2003794117

DO - 10.1073/pnas.2003794117

M3 - ???researchoutput.researchoutputtypes.contributiontojournal.article???

C2 - 32873650

AN - SCOPUS:85091072990

SN - 0027-8424

VL - 117

SP - 22743

EP - 22751

JO - Proceedings of the National Academy of Sciences of the United States of America

JF - Proceedings of the National Academy of Sciences of the United States of America

IS - 37

ER -

Restoration of fragmentary Babylonian texts using recurrent neural networks

Abstract

Keywords

Access to Document

Other files and links

Fingerprint

Cite this