PhD position Digital Humanities, AI, and Citizen Science

at  Universiteit Twente

KEY TAKEAWAYS

• 
  Hours
  38 - 40 hr.
• 
  Salary indication
  Salary gross/monthly
  based on full-time
  € 2,872 - € 3,670
• 
  Deadline
  13 Dec 2024
  We are looking for a talented and motivated PhD candidate in the field of Digital Humanities to research the motivation of citizen scientists to annotate textual and graphic patterns in digitized multimodal collections (e.g. handwritten archives). The candidate should also be able to evaluate and co-develop a web interface that ensures the appropriate division of labour between ‘humans in the loop’ and machines in the context of continuously learning Artificial Intelligence (AI)-driven systems.
  Your research will help to create innovative feedback loops between engaged volunteers and machine-generated output: recurring textual/graphic patterns (e.g. headings, tables, marginalia) or visualizations of basic narrative components (e.g. place, circumstances of a historical event, thematic aspects) as can be found in many multimodal archive collections. Although there have been enormous developments in the field of automated handwriting recognition (HTR), problems at the level of layout analysis still need to be addressed . In order to exploit deep learning in these areas, it is necessary to have ‘humans/citizens in the loop’, who not only label characters and words and transcribe lines but also train machines to analyse images in a multifaceted manner.
  Your work – together with the work of other researchers in Work Package 2 of the HAICu project (https://www.haicu.science/) – will contribute to new ways of multimodal data mining on the basis of layout and graphical attributes. As a PhD researcher, you will work within the HAICu research project. The HAICu (digital Humanities - Artificial Intelligence - Cultural heritage) project is a large-scale Dutch research project jointly pursued by universities and cultural-heritage institutions. We study new forms of Artificial Intelligence-based access to multimodal cultural-heritage data, both contemporary and historical. Within HAICu, Digital Humanities researchers, AI researchers, and a wide range of public and private partners will co-develop scientific solutions to unlock the true societal potential of heterogeneous digital heritage collections. It will provide easier, richer, and more reliable data access to citizens, journalists, civic organisations, and various other stakeholders.
  Within the HAICu team, you will participate in Work Package 2, titled Continual Machine Learning and Humans in the Loop. In this Work Package, we will collaborate with AI researchers from the University of Groningen and NHL Stenden, and partners from Dutch cultural heritage institutions, such as the Nationaal Archief (Dutch National Archives), Collectie Overijssel, and Groninger Archieven. As a PhD student in the TPS department, you will collaborate with a team of supervisors, including Dr. Andreas Weber, Dr. Annemieke Romein, and Dr. Koray Karaca. At the UT, you will work in a vibrant environment with diverse research projects and opportunities for collaboration and exchange.
  In addition to publishing scientific articles, the researcher is expected to make code available to academic and institutional project partners under open-access conditions.
  For more information about HAICu, please see https://www.haicu.science/

YOUR PROFILE

We are looking for a talented and motivated researcher who:

• has or will shortly acquire a (research) master’s degree or equivalent in digital humanities and/or neighbouring fields (for example, science and technology studies, digital history, philosophy of technology, computer science, psychology, human media interaction)
• is motivated to pursue independent research and has the willingness to develop writing and publication skills
• has excellent communication skills, works well in a team and enjoys collaboration
• has basic programming skills and hands-on experience with evaluation of web (annotation) interfaces; demonstrated expertise in UX design is of advantage
• has the skills and motivation to design and organize (together with cultural heritage stakeholders) sessions in which volunteers experiment with new forms of annotation
• fluent proficiency in English (written and oral) and, preferably, a good command of Dutch is an advantage.

Please refer the Job description for details