Creating, Enriching and Valorizing Treebanks of Ancient Greek

Keersmaekers, Alek; Mercelis, Wouter; Swaelens, Colin; Hal, Toon Van

doi:10.18653/v1/w19-7812

Cited by 9 publications

(5 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Such a pre-processing pipeline consists of three parts: a tokenizer to split the words and punctuation, a part-of-speech tagger to perform morphological analysis and a lemmatiser to provide every word with its lemma. Such pipelines already exist for ancient Greek (Crane 1991, Keersmaekers et al 2019) but those are dictionary-based approaches that cannot deal with inconsistent orthography or out-ofvocabulary words. For example, a dictionary-based approach will be able to analyse βιβλίου as a genitive from βιβλίον ("book"), but the variant βοιβληου, as it occurs in Occurrence 32232, will not be recognised as such.…”

Section: Iii1 Nlp Subprojectmentioning

confidence: 99%

The Database of Byzantine Book Epigrams Project: Principles, Challenges, Opportunities

Ricceri

Bentein

Bernard

et al. 2023

Journal of Data Mining &Amp; Digital Humanities

View full text Add to dashboard Cite

This paper presents an overview of the history, conceptualization, and development of the Database of Byzantine Book Epigrams, an ongoing research project hosted at Ghent University. It also offers a glimpse into current and future research threads carried out within the project, with an eye on long-term sustainability. The first part of the paper pinpoints the position of DBBE within the broad field of Digital Humanities and addresses the question of how and why Byzantine metrical paratexts have been collected in an open-access online database. In the second part of the article, we describe the main features of the relational database currently available, both from the perspective of its users and from a technical point of view. The third section of the paper includes the description of four subprojects connected to DBBE, which at present involve the development of a graph database complementary to the relational one, the implementation of natural language pre-processing applied to the DBBE corpus, the linguistic analysis of formulaicity in book epigrams, and the exploration of the broad implications of the study of book epigrams for a better understanding of Byzantine book culture.

show abstract

Section: Iii1 Nlp Subprojectmentioning

confidence: 99%

The Database of Byzantine Book Epigrams Project: Principles, Challenges, Opportunities

Ricceri

Bentein

Bernard

et al. 2023

Journal of Data Mining &Amp; Digital Humanities

View full text Add to dashboard Cite

show abstract

“…ongoing dependency treebank initiatives for Ancient Greek will be briefly described here (see also Keersmaekers et al 2019). The two most extensive projects that exist to date are the Perseus Ancient Greek (and Latin) Dependency Treebanks (AGDT) (Bamman, Mambrini, and Crane 2009) and the PROIEL Treebank (Haug and Jøhndal 2008).…”

Section: State Of the Field And Related Workmentioning

confidence: 99%

“…It is also worthwhile to mention the Harrington Trees, containing -among other texts -Lucian's True Histories (Harrington 2018). Finally, for this project, we developed our own collection of syntactic trees, the Pedalion Treebanks (Keersmaekers et al 2019), including classical and post-classical prose and poetry, with a special focus on genres and authors that are less well represented in the major treebanking projects -see Sections 4.1 and 6 for more details. This outline shows that, until now, there has been a strong emphasis on setting up treebank projects and establishing annotation conventions.…”

Section: State Of the Field And Related Workmentioning

confidence: 99%

Creating a large-scale diachronic corpus resource: Automated parsing in the Greek papyri (and beyond)

Keersmaekers

Hal

2023

Nat. Lang. Eng.

View full text Add to dashboard Cite

This paper explores how to syntactically parse Ancient Greek texts automatically and maps ways of fruitfully employing the results of such an automated analysis. Special attention is given to documentary papyrus texts, a large diachronic corpus of non-literary Greek, which presents a unique set of challenges to tackle. By making use of the Stanford Graph-Based Neural Dependency Parser, we show that through careful curation of the parsing data and several manipulation strategies, it is possible to achieve an Labeled Attachment Score of about 0.85 for this corpus. We also explain how the data can be converted back to its original (Ancient Greek Dependency Treebanks) format. We describe the results of several tests we have carried out to improve parsing results, with special attention paid to the impact of the annotation format on parser achievements. In addition, we offer a detailed qualitative analysis of the remaining errors, including possible ways to solve them. Moreover, the paper gives an overview of the valorisation possibilities of an automatically annotated corpus of Ancient Greek texts in the fields of linguistics, language education and humanities studies in general. The concluding section critically analyses the remaining difficulties and outlines avenues to further improve the parsing quality and the ensuing practical applications.

show abstract

“…While the projects mentioned above only include the full text, there have also been some efforts to add linguistic annotation. A wide variety of treebanking projects have manually annotated Greek texts for morphology, lemmas, (dependency) syntax and sometimes semantics, most prominently the PROIEL project (Haug and Jøhndal, 2008; 277,000 tokens), the Ancient Greek Dependency Treebanks (AGDT; Bamman et al, 2009; 560,000 tokens), the Gorman trees (Gorman, 2020; 324,000 tokens) and the Pedalion project (Keersmaekers et al, 2019; 320,000 tokens), as well as some smaller projects (in total, the manually annotated work includes about 1.5 million tokens). The former two projects are also included in the Universal Dependencies (UD) project (Nivre et al, 2016).…”

Section: Related Workmentioning

confidence: 99%

Proceedings of the 2nd International Workshop on Computational Approaches to Historical Language Change 2021

2021

View full text Add to dashboard Cite

Linguistic diversity as a testing ground for the study of semantic change Abstract: There are between 6000 and 8000 languages currently spoken in the world. The majority of those still lack decent descriptions, not to mention any written tradition and sizeable documents to rely on while trying to trace semantic changes they have undergone in the past and understanding the mechanisms behind them. Understandably, but likewise regrettably, most of the theoretical thinking in linguistics and adjacent disciplines has been formed by research on a few very big languages with a long written tradition, and the same has to a large extent been carried over to computational approaches, including work on semantic change. In our talk we will focus on two big issues which we believe deserve more awareness and attention among researchers involved in computational approaches to historical language change:• A crucial part in any theoretical work consists of formulating hypotheses, generalizations, laws etc. and explaining them, and work on semantic change is, of course, no exception. Linguistic diversity does not imply that any such generalizations are meaningless or premature before these have been studied for all the world's languages. It does imply, though, that such generalizations gain a lot from careful systematic cross-linguistic research that may unveil cross-linguistic regularities behind diversity -which is foundational for linguistic typology. Here we will discuss several cases whereby such research has questioned earlier generalizations on semantic change based on the familiar languages and/or has come up with new hypotheses.• But given that the majority of the world's languages lack any written tradition and sizeable historical documents, how is it possible to study semantic changes they have undergone in the past? This is indeed a big challenge, but not an insurmountable one. We will discuss several methods which often combine a careful intragenetic comparison (i.e., comparison of closely related languages) and a broader cross-linguistic perspective and some of the results obtained by their application.

show abstract

Creating, Enriching and Valorizing Treebanks of Ancient Greek

Cited by 9 publications

References 10 publications

The Database of Byzantine Book Epigrams Project: Principles, Challenges, Opportunities

The Database of Byzantine Book Epigrams Project: Principles, Challenges, Opportunities

Creating a large-scale diachronic corpus resource: Automated parsing in the Greek papyri (and beyond)

Proceedings of the 2nd International Workshop on Computational Approaches to Historical Language Change 2021

Contact Info

Product

Resources

About