Ithaca: Artificial Intelligence Tool Helps Restoration and Attribution of Ancient Texts | Science

The inscriptions, texts written on durable materials, can be a very good source for learning about the life and society of the time. One of the problems historians encounter is that these inscriptions are sometimes incomplete or have been moved from the place where they were created. In 2017, Yannis Assael, artificial intelligence scientist and current Google DeepMind researcher, and Thea Sommerschield, historian and Marie Curie fellow at the University of Venice, discussed these challenging tasks for historians and concluded the great cooperative potential between its disciplines. This is how Ithaca was born, a deep neural network for textual restoration and geographic and chronological attribution of ancient Greek inscriptions. This tool is designed to work together with history professionals: by itself it achieves an accuracy of 62% to restore damaged texts, but together with the work of historians it manages to improve the accuracy of professionals from 25% to 72%. . The percentage of accuracy in the location is 71% and, in addition, it can date the inscriptions in less than 30 years.

In order to maximize collaboration between historians and deep learning, this tool offers multiple hypotheses. For restoration, Ithaca provides 20 predictions decoded and ranked by probability. In this way, it is easier for historians to choose between the suggestions of the tool, taking into account their knowledge. Regarding geographic attribution, the instrument classifies the results among 84 regions; the list of candidate regions is implemented in a map and bar chart. Finally, instead of offering a value for chronology, it predicts a categorical distribution over dates, which are grouped in 10-year intervals, between 800 BC and 800 AD.

“Our goal was to explore how machine learning can help historians better interpret these inscriptions, providing a richer understanding of ancient history,” comment Assael and Sommerschield. The name of the tool was chosen because it refers to the Greek island in Homer’s Odyssey. “We think it would be a good name because it has the meaning of returning something or someone to their origins,” they detail.

Ithaca is a type of artificial intelligence called a deep learning model (deep learning model). It is based on a neural network, inspired by the neural networks found in the human brain, according to the researchers. “We’ve trained computers to use these neural networks for large amounts of data, so they can apply what they’ve learned to new data they haven’t seen before. It is trained on the largest dataset of ancient Greek inscriptions. Another aspect that stands out is that, due to the architecture of the tool, it is easily applicable to any ancient language, “from Latin to Mayan and Arcadian”, in addition to any written medium, “from papyri to manuscripts”. Ithaca, the result of multidisciplinary research by different entities and companies, has been featured in a study recently published in Nature Y andIt is publicly available on its website.

Fernando Notario, doctor in History, professor of Geography and History and whose line of research, among others, is the historiography of classical antiquity, summarizes that the main problems that professionals have in this specific discipline are the sources because a minimal fraction of documents from the ancient world. “It is not a magical solution, but it is a way to streamline and “objectify” things that we are already doing recently,” he says. Nerea Luis, doctor in Artificial Intelligence, defends that it makes “a lot of sense” to do it collaboratively with artificial intelligence and that this collaboration model “is going to be seen a lot. “It also leads you not to forget.”

To prevent some languages ​​from being forgotten, the Massachusetts Institute of Technology (MIT) presented in 2020 a system developed by MIT Computer Science and Artificial Intelligence (CSAIL) researchers capable of automatically deciphering a lost language, without the need for advanced knowledge of its relationship with other languages. In addition, it can determine relationships between languages ​​on its own. The ultimate goal is that the system can help linguists to decipher languages ​​that have been lost throughout history.

You can follow MATTER in Facebook, Twitter and Instagramor sign up here to receive our weekly newsletter.