PLOS ONE RESEARCH ARTICLE Reading Akkadian cuneiform using natural language processing 1 2☯ 2☯ 3 4 Shai GordinID *, Gai Gutherz , Ariel Elazary , Avital Romach , Enrique JimeÂnez , Jonathan Berant2, Yoram Cohen3 1 Faculty of Social Sciences and Humanities, Digital Humanities Ariel Lab, Ariel University, Ariel, Israel, 2 School of Computer Sciences, Tel Aviv University, Tel Aviv, Israel, 3 Jacob M. Alkow Department of Archaeology and Ancient Near Eastern Civilizations, Tel Aviv University, Tel Aviv, Israel, 4 Institute for Assyriology and Hittitology, Ludwig-Maximilians-UniversitaÈt MuÈnchen, Munich, Germany ☯ These authors contributed equally to this work. a1111111111 *
[email protected] a1111111111 a1111111111 a1111111111 Abstract a1111111111 In this paper we present a new method for automatic transliteration and segmentation of Unicode cuneiform glyphs using Natural Language Processing (NLP) techniques. Cunei- form is one of the earliest known writing system in the world, which documents millennia of OPEN ACCESS human civilizations in the ancient Near East. Hundreds of thousands of cuneiform texts Citation: Gordin S, Gutherz G, Elazary A, Romach were found in the nineteenth and twentieth centuries CE, most of which are written in Akka- A, JimeÂnez E, Berant J, et al. (2020) Reading dian. However, there are still tens of thousands of texts to be published. We use models Akkadian cuneiform using natural language based on machine learning algorithms such as recurrent neural networks (RNN) with an processing. PLoS ONE 15(10): e0240511. https:// accuracy reaching up to 97% for automatically transliterating and segmenting standard Uni- doi.org/10.1371/journal.pone.0240511 code cuneiform glyphs into words. Therefore, our method and results form a major step Editor: Marco Lippi, University of Modena and towards creating a human-machine interface for creating digitized editions.