Restoring ancient text using deep learning: a case study on Greek epigraphy

Proceedings of the 2019 Conference on Empirical Methods in Natural Language Processing and the 9th International Joint Conference on Natural Language Processing, pages 6368–6375, Hong Kong, China, November 3–7, 2019. c⃝2019 Association for Computational Linguistics 6368 Restoring ancient text using deep learning: a case study on Greek epigraphy Yannis Assael1,2,†, Thea Sommerschield1,† , Jonathan Prag1 University of Oxford1, DeepMind2 Abstract Ancient History relies on disciplines such as Epigraphy, the study of ancient inscribed texts, for evidence of the recorded past. However, these texts, “inscriptions”, are often damaged over the centuries, and illegible parts of the text must be restored by specialists, known as epigraphists. This work presents PYTHIA, the first ancient text restoration model that recov- ers missing characters from a damaged text input using deep neural networks. Its archi- tecture is carefully designed to handle long- term context information, and deal efficiently with missing or corrupted character and word representations. To train it, we wrote a non- trivial pipeline to convert PHI, the largest digi- tal corpus of ancient Greek inscriptions, to ma- chine actionable text, which we call PHI-ML. On PHI-ML, PYTHIA’s predictions achieve a 30.1% character error rate, compared to the 57.3% of human epigraphists. Moreover, in 73.5% of cases the ground-truth sequence was among the Top-20 hypotheses of PYTHIA, which effectively demonstrates the impact of this assistive method on the field of digital epigraphy, and sets the state-of-the-art in an- cient text restoration. 1 Introduction One of the key sources for Ancient History is the discipline of epigraphy, which delivers firsthand evidence for the thought, society and history of an- cient civilisations. Epigraphy is the study of doc- uments, “inscriptions”, written on a durable sur- face (stone, ceramic, metal) by individuals, groups and institutions of the past (Davies and Wilkes, 2012). Only a small minority of surviving inscrip- tions are fully legible and complete, as many have been damaged in time (Figure 1). An epigraphist must then hypothesise how much text is missing, †These authors contributed equally to this work. and what it might have originally been. These hy- potheses are called “restorations” (Bodel, 2012). The present work offers a fully automated aid to the epigraphist’s restoration task. taskRestoring text is a complex and time-consuming (Woodhead, 1967; Mattingly, 1996). Epigraphists rely on accessing vast repositories of information to find textual and contextual “parallels” (recurring expressions in similar documents). These repositories primarily consist in a researcher’s mnemonic repertoire of such parallels, and in digital corpora for performing “string matching” searches (The Packard Human- ities Institute, 2005; Clauss, 2012). However, minor differences in the search query can exclude or obfuscate relevant results, making it hard to estimate the true probability distribution of pos- sible restorations. To the best of our knowledge, this is the first work to bypass the constraints of current epigraphic methods by means of a fully automated deep learning model, PYTHIA, which aids the task of ancient text restoration. It is supplemented by PHI-ML, an epigraphic dataset of a machine actionable text. PYTHIA takes as …

Content truncated to 500 words