Recent Advances in Digital Humanities

The three sections of the conference are reflected in the topic organization of this volume: Resources and digitalization, Tools and interfaces, and Computational methods analysis, classification, and clustering. The section order reflects the natural lifespan of a digital object, from its creation and representation to its pre-processing and processing and, finally, to its analysis and interpretation.

2. Content: Sections and chapters overview

During the pandemic, many scholars have turned to digital resources and have expanded the use of digital tools to several areas of research and communication. At the time when mobility restrictions limited the physical space one could access, the virtual world was always available. Social media, conference platforms, eLearning systems, and digital repositories or libraries were quickly adopted by scholars.

The first section of the book, Resources and Digitalisation has an applicative character. It starts from the awareness that Digital Humanities has become the buzzword in academic circles and beyond (Burghardt and Wolff 2017). More and more research initiatives approach traditional or new emerging topics related to the field of Digital Humanities for higher visibility and faster contribution to scientific development. But such effervescence is challenging to grasp; that is why it is important to obtain latest information on existing resources and digitalization strategies. At the same time, it is essential to acknowledge emerging synergies between the disciplines that can support further advancements for the larger Digital Humanities domain. One such complementarity relationship has been shaped around the discipline of corpus linguistics. These two aspects represent the focus in Section 1 (five chapters). The topic is also adapted for the target languages of the volume, namely Romance languages. Contributions on this topic showcase the process of using corpus linguistics methods to teach and conduct teaching-oriented research in the humanities. Exemplifications of typical tools (for example, OCR techniques for Italian documents) for text digitalizations are also provided.

Chapter 2, “Phraseology in Romanian Academic Writing: Corpus Based Explorations into Field-Specific Multiword Units”, by Valentina Mureșan, Roxana Rogobete, Ana-Maria Bucur, Mădălina Chitez and Andreea Dincă, draws on a comparative analysis of two self-complied corpora (ROGER Corpus of Romanian Academic Genres and EXPRES Corpus of Expert Writing in Romanian and English) which are used to test a computational method for extracting filed-specific multi-word units in Romanian academic writing. The findings of the study serve as a first stage in the development of a collocation or n-gram extraction method which results in the creation of academic word / collocation / phrase lists for the Romanian language.

Tommaso Spinelli’s research (Chapter 3), “The Latin Diachronic Database: A New Digital Tool for the Study of Latin”, highlights a new digital toolkit that helps scholars and students to identify and analyse the use of Latin in literary texts, both from the perspective of the words’ frequency and development throughout time.

In Chapter 4, “A Proposal for a Multilingual E-Glossary of Discourse Markers”, Cecilia Mihaela Popescu and Oana Adriana Duță aim to establish an e-glossary of discourse markers in English and four romance languages, with the help of digital tools; this process would help scholars in their linguistic research endeavours.

Section 1 is closed by Eugen Istodor’s chapter (Chapter 5): “The Meme That Brings About the Roar. Then, Discredit. The Tismăneanu Case in Pandemic Times”, which offers a critical analysis of how social codes and social media have been reshaped in the early pandemics. Istodor examines the case of Vladimir Tismăneanu, professor of political sciences in the United States, and an important public figure in the Romanian cultural space. The use of social media, and especially the culture of sharing encouraged by social media platforms, brings into question the relation between private and public space, but also the shifted boundaries between humour and discrimination in the case of a meme shared in the early pandemics.

Because the electronic resources such as corpora, archives, repositories, etc. exploded in the last decade and the data quantity generated by humanity on a daily basis is overwhelming, managing this kind of data by hand or with only a few specialized professionals has become simply unfeasible. The sheer quantity and complexity of data nowadays require more and more user-friendly and intelligent tools and platforms to be processed with. Almost in every domain of the Humanities, the apparition of such new tools and interfaces is a reality. For instance, in disciplines such as the history of philosophy and science, which rely more on traditional library and archival research, the use of digital resources was, during the pandemic, the predominant way of doing research. The four chapters in this section are a fair sample of that research phenomenon.

The section begins with Chapter 6, “Natural Language Processing for Book Recommender Systems”, by Haifa Alharthi and Diana Inkpen, who describe several approaches to a literary book recommender system, that takes the text of the books into consideration. One approach considers the writing style of the authors by transferring the information learned by an author-identification model into a book recommender system. Another approach represents books using more than one hundred linguistic features including lexical, syntactic, stylometric, and fiction-based features. The two proposed systems are evaluated in a top-k recommendation scenario, and both provide higher recommendation accuracy compared with state-of-the-art content and collaborative filtering methods.

Chapter 7, “Syntactic Tree Editor”, authored by Claudius-Marian Teodorescu, presents the development of an editor for syntactic trees, to be used during the process of teaching grammar in universities or for research purposes. It complies with the Government and Binding (GB) model, but there is the possibility of extending it to support various other grammar models. The editor was designed after a comprehensive analysis of the existing software products for editing or generating syntax trees. The functionality of the tree editor and its user-friendly interface are presented in detail.

The last chapter of the section (Chapter 9), “An Evaluation of the Ithaca Tool Performance for Restoring Lost Texts (Ancient Greek)”, co-authored by Alexandra Lițu and Valentin Bottez, aims to evaluate the performance of the Ithaca tool on a funerary inscription, coming most probably from Callatis, a Greek colony on the Western shore of the Black Sea, that was fully preserved and published recently by the authors. To overcome the requirement of having at least 50 characters in order to use Ithaca, they performed some text augmentation. However, these augmentations might have proved influential for geographical and chronological attribution. One of the salient features of Ithaca seems to be its receptiveness to context. It also proved itself very sensitive to slight changes in the input text, even on the level of basic punctuation.