CROSS 2021

Theme: Digital Humanities

The dynamic and fast-growing field of digital humanities brings research in the arts, social sciences, and humanities together with cutting-edge empirical methods from computer science, data science, and engineering.

The digital humanities link domain knowledge with digital and computational tools to analyze, model, and critically reflect on social and cultural issues and real-world challenges. Researchers in the field both apply computational methods to the humanities, and integrate humanities thinking into digital systems. Common areas of application include, but are not limited to: cultural heritage, historical archives, audiovisual archives, social media interaction, literature, musicology, user experience, museology, linguistics, humanitarian work, etc.

CROSS 2021: Selected projects

Aris Xanthos UNIL (SLI FL)
Robert West EPFL (IC DLAB)
Anita Auer UNIL (SLI FL)

Among new ways of expression emerging in computer-mediated communication (CMC), emojis have become extremely popular worldwide, particularly in interpersonal conversations. The ACCOMOJI project seeks to examine the ways in which people conversing in the Swiss national languages converge or diverge over time with regard to emoji usage, thereby managing social and emotional distance. Based on the What’s Up Switzerland (WUS) corpus of Whatsapp conversations, a citizen science approach will be taken to annotate emojis in terms of function and emotional content. Data science methods will then capture accommodation patterns in the annotated data and correlate them with demographic features. Besides an assessment of the appropriateness of various formal models and methods for addressing emotional aspects of interpersonal CMC, the project will result in a better understanding of citizen science practices and their applicability to language-related research in the Swiss multilingual context.

François Bavaud UNIL (SLI FL)

The “dualism debate”, a hot topic in 19th-century German music theory, is concerned with the mutual relationship of major and minor triads, specifically whether the minor triad is a mere derivative of the major triad or whether it can be derived from first principles on its own right. This project strives to reconstruct and critically evaluate the discursive relations within this debate by using the combined power of qualitative-historical and quantitative-numerical methods and thus serve to build methodological bridges between the humanities and the sciences. The main contribution of this project consists in an empirical reconstruction of the discursive structure of the dualism debate using network and topic modeling. The project will lead to a publicly-available online resource that displays the results of the research project and provides access to the data for future research.

Davide Picca UNIL (SLI FL)
Sarah Kenderdine EPFL (CDH DHI eM+)

Since UNESCO affirmed the importance of intangible heritage in 2003, efforts to model folklore, tradition, experience and other forms of cultural heritage have been sparse. The primary objective of this project is to explore the potential for standardizing the consumption of tangible and intangible cultural heritage by instantiating a use case in intangible heritage for the Hong Kong Martial Arts Living Archive (a collaboration between EPFL’s Laboratory for Experimental Museology and the International Goushou Association). It will do so by developing an ontological model of intangible heritage combining aspects of haptics, pose and motion with traits of tradition and folklore, and generating a linked knowledge graph for Kung Fu heritage. The resulting data source will then be integrated into the Digital Humanities Toolkit (DHTK), a software library developed by SLI-UNIL that enables unified, systematic access to and consumption of online digital collections. It will set a methodological framework for integrated tangible and intangible heritage.

Marie-Hélène CÔTE, UNIL (SLI FL)

Recent progress in Digital Humanities opens new avenues for creating large databases of archival documents. The elaboration of these databases relies on advanced optical character recognition techniques, working now for both printed and handwritten writing. The evolution of a population can now be reconstructed through the systematic collection of ancient administrative documents, for the benefit of a variety of historical studies. The project “Names of Lausanne” aims to develop a database of the population of Lausanne between 1803 and 1900 through the automatic extraction of archival documents from different heritage institutions in Lausanne (civil records, census data, directories). This database will be first exploited for linguistic analysis, in particular regarding the evolution of name variants.