The project Pop-Lite: Romanian Popular Novels and Their Sub-genres during the Long Nineteenth Century: digital editing and corpus analysis aimed to remedy a number of deficits and gaps in the field of Romanian computational literary studies (CLS), assuming the following objectives:
a. the realization of digitized resources comprising mostly non-canonical literary works (100% achieved);
b. the constitution of literary corpora starting from the operationalization of some concepts of literary theory (100%);
c. the application of processing and analysis tools and their testing on samples of modern literary Romanian language (100%).
d. engaging a critical mass of researchers in an interdisciplinary investigation of data extracted from literary sources, focusing on the theoretical, methodological and practical effects of distant reading in Romanian literary history and theory (100%).

The total number of results is: 4 articles indexed in ISI/ BDI; 17 conference presentations; 4 chapters in volumes; 1 volume; 23 “other results” (including the Pop-Lite Corpus, as well as the documentation detailing the project implementation methodology). At the same time, various partial results (repositories) were realized, containing either pdf files, or various forms through which editable files have passed, or various file formats (xml, txt), or datasets useful for future quantitative analysis (stylometry, discourse analysis). Accordingly, the contractual conditions are fulfilled.

Of the results, the Pop-Lite corpus will have the greatest impact. We support our claim on the following arguments:

• it is the only literary resource in Romanian automatically annotated with POS, MSD and NER (models available in 2024);
• it is the only corpus that illustrates the phenomenon of the popular novel/consumer literature;
• it is a dynamic and reusable resource, specifically designed to be modeled according to users’ research interests;
• it is a stable resource both in the platforms through which we make it available (Zenodo, Github, TextGrid etc.) and in the data format (TEI-xml);
• it is a metadata resource linked to a widely accepted ontology (https://opentheso.hum num.fr/opentheso/api/theso/Typology).

In its current version, the Pop-Lite Corpus contains 100 novels + an extension of 23 novels. The extension contains novels that have either been included in other corpora (ELTEC, HAIRO) or have a proofreading/transcription accuracy of less than 80%. The repository on Github has 2 folders with unannotated TEI-XML files, 2 folders with annotated TEI-XML files (POS+MSD+NER), 1 folder with scripts and 1 folder with tests. While the Github format is dynamic (being open access data, user input is encouraged), the Zenodo format contains the annotated POS, MSD and NER files at the performance level that the available models could reach (https://zenodo.org/records/13771810).

Traditional theoretical issues such as the literary canon, periodization, authorship attribution and literary movements can be redefined through a new perspective supported by corpora of popular (consumer) literature, nineteenth-century contexts and Computational Literary Studies. The most effective way to overcome the theoretical difficulties encountered so far in the research of modern Romanian texts would be an initiative that combines the tools and methods of Digital Humanities with the tradition of Romanian literary history and theory, which should by no means be ignored or disregarded. Pop-Lite is the first Romanian research project to propose such an approach.

The full final report is available here.

Citation suggestion

Patras, R., Pascariu, L., Cuzmici, L., Olteanu, A., Ionescu, O., & Bădeliță, T. (2024). Corpus Pop-Lite [Data set]. Zenodo. https://doi.org/10.5281/zenodo.13771810

The corpus can be accessed through Zenodo și Github.

Corpus Pop-Lite

EXTENSIE CORPUS Pop-Lite

PROJECT OBJECTIVES

• O1: to create a literary reference corpus for the study of the 19th-century popular novel and its sub-genres;
• O2: to test and customize existing analysis tools in DLS (digital literary studies) on the diachronic varieties of the Romanian language;
• O3: development of young researchers’ skills in DLS and DR (distant reading), including the assimilation of digital skills, best practices, and methods;
• O4: engaging a critical mass of researchers in an interdisciplinary investigation of data extracted from literary sources, focusing on the theoretical, methodological, and practical effects of the DR approach in Romanian literary history and theory.

RESEARCH TEAM

RESULTS

1. DHL best practice documentation (guidelines, reports, white papers, workflow);
2. 3 articles/chapters in volumes and 1 volume; 
3. 5 conference papers; 
4. 8 mobility reports; 
5. 1 corpus in open access; 
6. 1 shadow repository; 
7. 1 website; 
8. 2 workshops.