Restoring and attributing ancient texts using deep neural networks. Assael, Y., Sommerschield, T., Shillingford, B., Bordbar, M., Pavlopoulos, J., Chatzipanagiotou, M., Androutsopoulos, I., Prag, J., & de Freitas, N. Nature, 603(7900):280–283, March, 2022. 🏷️ /unread、Archaeology、History、Computer science![Restoring and attributing ancient texts using deep neural networks [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper doi abstract bibtex Ancient history relies on disciplines such as epigraphy—the study of inscribed texts known as inscriptions—for evidence of the thought, language, society and history of past civilizations1. However, over the centuries, many inscriptions have been damaged to the point of illegibility, transported far from their original location and their date of writing is steeped in uncertainty. Here we present Ithaca, a deep neural network for the textual restoration, geographical attribution and chronological attribution of ancient Greek inscriptions. Ithaca is designed to assist and expand the historian’s workflow. The architecture of Ithaca focuses on collaboration, decision support and interpretability. While Ithaca alone achieves 62% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25% to 72%, confirming the synergistic effect of this research tool. Ithaca can attribute inscriptions to their original location with an accuracy of 71% and can date them to less than 30 years of their ground-truth ranges, redating key texts of Classical Athens and contributing to topical debates in ancient history. This research shows how models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history. 【摘要翻译】古代历史依赖于书法等学科–对铭刻文本的研究，即碑文–来证明过去文明的思想、语言、社会和历史1。 然而，几个世纪以来，许多碑文已损坏到难以辨认的地步，被运离了原来的位置，其书写日期也充满了不确定性。在此，我们介绍一种深度神经网络 Ithaca，用于古希腊碑文的文本修复、地理归属和年代归属。伊萨卡旨在辅助和扩展历史学家的工作流程。Ithaca 的架构侧重于协作、决策支持和可解释性。在修复受损文本时，Ithaca 的单独准确率为 62%，而历史学家使用 Ithaca 后，准确率从 25% 提高到 72%，这证实了该研究工具的协同效应。Ithaca 能以 71% 的准确率将碑文归属到其原始位置，并能将其年代精确到其地面实况范围的 30 年以内，从而重新确定了雅典古典时期重要文献的年代，并为古代历史的热点辩论做出了贡献。这项研究表明，像 "伊萨卡 "这样的模型可以释放人工智能与历史学家之间的合作潜力，从而对我们研究和书写人类历史上最重要时期之一的方式产生变革性影响。
@article{assael2022,
title = {Restoring and attributing ancient texts using deep neural networks},
volume = {603},
copyright = {2022 The Author(s)},
issn = {0028-0836},
shorttitle = {利用深度神经网络还原古籍并确定其归属},
url = {https://www.nature.com/articles/s41586-022-04448-z},
doi = {10.1038/s41586-022-04448-z},
abstract = {Ancient history relies on disciplines such as epigraphy—the study of inscribed texts known as inscriptions—for evidence of the thought, language, society and history of past civilizations1. However, over the centuries, many inscriptions have been damaged to the point of illegibility, transported far from their original location and their date of writing is steeped in uncertainty. Here we present Ithaca, a deep neural network for the textual restoration, geographical attribution and chronological attribution of ancient Greek inscriptions. Ithaca is designed to assist and expand the historian’s workflow. The architecture of Ithaca focuses on collaboration, decision support and interpretability. While Ithaca alone achieves 62\% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25\% to 72\%, confirming the synergistic effect of this research tool. Ithaca can attribute inscriptions to their original location with an accuracy of 71\% and can date them to less than 30 years of their ground-truth ranges, redating key texts of Classical Athens and contributing to topical debates in ancient history. This research shows how models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history.
【摘要翻译】古代历史依赖于书法等学科--对铭刻文本的研究，即碑文--来证明过去文明的思想、语言、社会和历史1。 然而，几个世纪以来，许多碑文已损坏到难以辨认的地步，被运离了原来的位置，其书写日期也充满了不确定性。在此，我们介绍一种深度神经网络 Ithaca，用于古希腊碑文的文本修复、地理归属和年代归属。伊萨卡旨在辅助和扩展历史学家的工作流程。Ithaca 的架构侧重于协作、决策支持和可解释性。在修复受损文本时，Ithaca 的单独准确率为 62\%，而历史学家使用 Ithaca 后，准确率从 25\% 提高到 72\%，这证实了该研究工具的协同效应。Ithaca 能以 71\% 的准确率将碑文归属到其原始位置，并能将其年代精确到其地面实况范围的 30 年以内，从而重新确定了雅典古典时期重要文献的年代，并为古代历史的热点辩论做出了贡献。这项研究表明，像 "伊萨卡 "这样的模型可以释放人工智能与历史学家之间的合作潜力，从而对我们研究和书写人类历史上最重要时期之一的方式产生变革性影响。},
language = {en},
number = {7900},
urldate = {2022-09-28},
journal = {Nature},
author = {Assael, Yannis and Sommerschield, Thea and Shillingford, Brendan and Bordbar, Mahyar and Pavlopoulos, John and Chatzipanagiotou, Marita and Androutsopoulos, Ion and Prag, Jonathan and de Freitas, Nando},
month = mar,
year = {2022},
note = {🏷️ /unread、Archaeology、History、Computer science},
keywords = {/unread, Archaeology, Computer science, History},
pages = {280--283},
}
Downloads: 0
{"_id":"EHHbMYx986eykoDAn","bibbaseid":"assael-sommerschield-shillingford-bordbar-pavlopoulos-chatzipanagiotou-androutsopoulos-prag-etal-restoringandattributingancienttextsusingdeepneuralnetworks-2022","author_short":["Assael, Y.","Sommerschield, T.","Shillingford, B.","Bordbar, M.","Pavlopoulos, J.","Chatzipanagiotou, M.","Androutsopoulos, I.","Prag, J.","de Freitas, N."],"bibdata":{"bibtype":"article","type":"article","title":"Restoring and attributing ancient texts using deep neural networks","volume":"603","copyright":"2022 The Author(s)","issn":"0028-0836","shorttitle":"利用深度神经网络还原古籍并确定其归属","url":"https://www.nature.com/articles/s41586-022-04448-z","doi":"10.1038/s41586-022-04448-z","abstract":"Ancient history relies on disciplines such as epigraphy—the study of inscribed texts known as inscriptions—for evidence of the thought, language, society and history of past civilizations1. However, over the centuries, many inscriptions have been damaged to the point of illegibility, transported far from their original location and their date of writing is steeped in uncertainty. Here we present Ithaca, a deep neural network for the textual restoration, geographical attribution and chronological attribution of ancient Greek inscriptions. Ithaca is designed to assist and expand the historian’s workflow. The architecture of Ithaca focuses on collaboration, decision support and interpretability. While Ithaca alone achieves 62% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25% to 72%, confirming the synergistic effect of this research tool. Ithaca can attribute inscriptions to their original location with an accuracy of 71% and can date them to less than 30 years of their ground-truth ranges, redating key texts of Classical Athens and contributing to topical debates in ancient history. This research shows how models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history. 【摘要翻译】古代历史依赖于书法等学科–对铭刻文本的研究，即碑文–来证明过去文明的思想、语言、社会和历史1。 然而，几个世纪以来，许多碑文已损坏到难以辨认的地步，被运离了原来的位置，其书写日期也充满了不确定性。在此，我们介绍一种深度神经网络 Ithaca，用于古希腊碑文的文本修复、地理归属和年代归属。伊萨卡旨在辅助和扩展历史学家的工作流程。Ithaca 的架构侧重于协作、决策支持和可解释性。在修复受损文本时，Ithaca 的单独准确率为 62%，而历史学家使用 Ithaca 后，准确率从 25% 提高到 72%，这证实了该研究工具的协同效应。Ithaca 能以 71% 的准确率将碑文归属到其原始位置，并能将其年代精确到其地面实况范围的 30 年以内，从而重新确定了雅典古典时期重要文献的年代，并为古代历史的热点辩论做出了贡献。这项研究表明，像 \"伊萨卡 \"这样的模型可以释放人工智能与历史学家之间的合作潜力，从而对我们研究和书写人类历史上最重要时期之一的方式产生变革性影响。","language":"en","number":"7900","urldate":"2022-09-28","journal":"Nature","author":[{"propositions":[],"lastnames":["Assael"],"firstnames":["Yannis"],"suffixes":[]},{"propositions":[],"lastnames":["Sommerschield"],"firstnames":["Thea"],"suffixes":[]},{"propositions":[],"lastnames":["Shillingford"],"firstnames":["Brendan"],"suffixes":[]},{"propositions":[],"lastnames":["Bordbar"],"firstnames":["Mahyar"],"suffixes":[]},{"propositions":[],"lastnames":["Pavlopoulos"],"firstnames":["John"],"suffixes":[]},{"propositions":[],"lastnames":["Chatzipanagiotou"],"firstnames":["Marita"],"suffixes":[]},{"propositions":[],"lastnames":["Androutsopoulos"],"firstnames":["Ion"],"suffixes":[]},{"propositions":[],"lastnames":["Prag"],"firstnames":["Jonathan"],"suffixes":[]},{"propositions":["de"],"lastnames":["Freitas"],"firstnames":["Nando"],"suffixes":[]}],"month":"March","year":"2022","note":"🏷️ /unread、Archaeology、History、Computer science","keywords":"/unread, Archaeology, Computer science, History","pages":"280–283","bibtex":"@article{assael2022,\n\ttitle = {Restoring and attributing ancient texts using deep neural networks},\n\tvolume = {603},\n\tcopyright = {2022 The Author(s)},\n\tissn = {0028-0836},\n\tshorttitle = {利用深度神经网络还原古籍并确定其归属},\n\turl = {https://www.nature.com/articles/s41586-022-04448-z},\n\tdoi = {10.1038/s41586-022-04448-z},\n\tabstract = {Ancient history relies on disciplines such as epigraphy—the study of inscribed texts known as inscriptions—for evidence of the thought, language, society and history of past civilizations1. However, over the centuries, many inscriptions have been damaged to the point of illegibility, transported far from their original location and their date of writing is steeped in uncertainty. Here we present Ithaca, a deep neural network for the textual restoration, geographical attribution and chronological attribution of ancient Greek inscriptions. Ithaca is designed to assist and expand the historian’s workflow. The architecture of Ithaca focuses on collaboration, decision support and interpretability. While Ithaca alone achieves 62\\% accuracy when restoring damaged texts, the use of Ithaca by historians improved their accuracy from 25\\% to 72\\%, confirming the synergistic effect of this research tool. Ithaca can attribute inscriptions to their original location with an accuracy of 71\\% and can date them to less than 30 years of their ground-truth ranges, redating key texts of Classical Athens and contributing to topical debates in ancient history. This research shows how models such as Ithaca can unlock the cooperative potential between artificial intelligence and historians, transformationally impacting the way that we study and write about one of the most important periods in human history.\n\n【摘要翻译】古代历史依赖于书法等学科--对铭刻文本的研究，即碑文--来证明过去文明的思想、语言、社会和历史1。 然而，几个世纪以来，许多碑文已损坏到难以辨认的地步，被运离了原来的位置，其书写日期也充满了不确定性。在此，我们介绍一种深度神经网络 Ithaca，用于古希腊碑文的文本修复、地理归属和年代归属。伊萨卡旨在辅助和扩展历史学家的工作流程。Ithaca 的架构侧重于协作、决策支持和可解释性。在修复受损文本时，Ithaca 的单独准确率为 62\\%，而历史学家使用 Ithaca 后，准确率从 25\\% 提高到 72\\%，这证实了该研究工具的协同效应。Ithaca 能以 71\\% 的准确率将碑文归属到其原始位置，并能将其年代精确到其地面实况范围的 30 年以内，从而重新确定了雅典古典时期重要文献的年代，并为古代历史的热点辩论做出了贡献。这项研究表明，像 \"伊萨卡 \"这样的模型可以释放人工智能与历史学家之间的合作潜力，从而对我们研究和书写人类历史上最重要时期之一的方式产生变革性影响。},\n\tlanguage = {en},\n\tnumber = {7900},\n\turldate = {2022-09-28},\n\tjournal = {Nature},\n\tauthor = {Assael, Yannis and Sommerschield, Thea and Shillingford, Brendan and Bordbar, Mahyar and Pavlopoulos, John and Chatzipanagiotou, Marita and Androutsopoulos, Ion and Prag, Jonathan and de Freitas, Nando},\n\tmonth = mar,\n\tyear = {2022},\n\tnote = {🏷️ /unread、Archaeology、History、Computer science},\n\tkeywords = {/unread, Archaeology, Computer science, History},\n\tpages = {280--283},\n}\n\n","author_short":["Assael, Y.","Sommerschield, T.","Shillingford, B.","Bordbar, M.","Pavlopoulos, J.","Chatzipanagiotou, M.","Androutsopoulos, I.","Prag, J.","de Freitas, N."],"key":"assael2022","id":"assael2022","bibbaseid":"assael-sommerschield-shillingford-bordbar-pavlopoulos-chatzipanagiotou-androutsopoulos-prag-etal-restoringandattributingancienttextsusingdeepneuralnetworks-2022","role":"author","urls":{"Paper":"https://www.nature.com/articles/s41586-022-04448-z"},"keyword":["/unread","Archaeology","Computer science","History"],"metadata":{"authorlinks":{}}},"bibtype":"article","biburl":"https://api.zotero.org/groups/2386895/collections/7PPRTB2H/items?format=bibtex&limit=100","dataSources":["JFDnASMkoQCjjGL8E","u8q5uny4m5jJL9RcX"],"keywords":["/unread","archaeology","computer science","history"],"search_terms":["restoring","attributing","ancient","texts","using","deep","neural","networks","assael","sommerschield","shillingford","bordbar","pavlopoulos","chatzipanagiotou","androutsopoulos","prag","de freitas"],"title":"Restoring and attributing ancient texts using deep neural networks","year":2022}