All Cambridge Digital Humanities courses

This in-person workshop will provide an accessible, non-technical introduction to Machine Learning systems, aimed primarily at graduate students and researchers in the humanities, arts and social sciences.

Key topics covered in the sessions will include:

• Situating Machine Learning in the longer history of Artificial Intelligence
• Machine Learning system architectures
• The challenges of dimension reduction, classification and generalisation
• Sources of bias and problems of interpretation
• Machine Learning applications and their societal consequences

During the session participants will be encouraged to work through practical exercises in image classification. No prior knowledge of programming is required. Participants wishing to run the experiments for themselves will need access to a laptop, but no special software is required, just an up-to-date web browser and an internet connection. We will be using Google Colab for the text generation experiments which you have access to via your Raven log-in. The image classification experiments will require a GitHub account ([sign up here https://github.com/])

Convenor: Estara Arrant (CDH Methods Fellow)

This methods workshop will teach students three powerful machine learning algorithms appropriate for Humanities research projects. These algorithms are designed to help you identify and explore meaningful patterns and correlations in your research material and are appropriate for descriptive, qualitative data sets of almost any size. These algorithms are applicable to virtually any Humanities field or research question.

• Multiple Correspondence Analysis: automatically identifies correlations and differences between specific data elements. This helps one to understand how different features (‘variables’ or ‘characteristics’) of one’s data are related to each other, and how strong their relationships are. This can be useful in almost any research project. For example, in a sociological dataset, this analysis could help clarify relationships between specific demographic characteristics (race, gender, political affiliation) and socioeconomic status (working class, education level, income bracket).
• K-modes clustering and hierarchical clustering: finding groups of similarity and relationship within the entirety of your data. Clustering helps one to identify which variables/characteristics group together, and which do not, and the degree of difference between groups. For example, such clustering could allow an art historian to see how paintings from one decade are characterised by style and artist, as contrasted to paintings from another decade (thus tracking shifts in artistic trends over time)

This workshop will specifically cover the following: Determining when your research could benefit from machine learning analysis. Designing a good methodology and running the analysis. Interpreting the results and determining if they are meaningful. Producing a useful visualisation (graphic) of the results. Communicating the findings to other scholars in the Humanities in an accessible way. Students will actively implement a small research project using a practice dataset and are encouraged to try out the methods in their current research. They will learn the basics of running the analysis in R’s powerful programming language.

This Methods Workshop explores primary data collection using digital and online qualitative methods. Teaching methods for detailed assessment of the suitability of online platforms for the collection of research data. Considering not only general ethical issues, privacy, encryption, terms and conditions but also inclusivity for neurodivergent and vulnerable participants.

Convenor: Orla Delaney (CDH Methods Fellow)

What does it mean to prioritise small data over big data?

Cultural heritage datasets, such as museum databases and digital archives, seem to resist the quantitative methods we usually associate with data science work, asking to be read and explored rather than aggregated and analysed. This workshop provides participants with a non-statistical toolkit that will enable them to approach, critique, and tell the story of a cultural heritage dataset.

Together we will consider approaches to the database from the history of science and technology, media archaeology, and digital ethnography. This will be done alongside an overview of practical considerations relevant to databasing in the sector, such as standards like FAIR (Findable, Accessible, Interoperable, Reusable) and CARE (Collective Benefit, Authority to Control, Responsibility, Ethics), specific technologies like linked data, and the results of recent projects aiming to criticise and diversify the underpinning technologies of cultural heritage databases. This workshop is aimed both at cultural heritage professionals and students, and at data science researchers interested in introducing a qualitative approach to their work.

This project begins from the premise that ‘transparency’ is not clear at all. Transparency is historically mediated, culturally constructed, and ideologically complex. Understood expansively, transparency is enmeshed with a variety of functions and associations, having been mobilised as a political call to action; a design methodology; a radical practice of digital disruption; an ideological tool of surveillance; a corporate strategy of diversion; an aesthetics of obfuscation; a cultural paradigm; a programming protocol; a celebration of Enlightenment rationality; a tactic for spatialising data; an antidote to computational black boxing; an ethical cliché; and more.

Across two workshops, we will explore the multidimensionality and intractability of transparency and investigate how the demand for more of it—in our algorithms, computational systems, and culture more broadly—can encode assumptions about the liberational capacity of restoring representation to the invisible. As a group we will conduct a survey of transparency and its political ramifications to digital culture by learning about its conceptual genealogies; interrogating its relevance to art and architecture; questioning its limits as an ethical imperative; and mapping it as a contemporary strategy of anti/mediation. Drawing on a combination of artworks, historical texts, cultural touchstones, and moving images, these workshops will give participants an opportunity to attend to transparency’s complex configurations within contemporary culture through a media theoretical lens. This project is designed to facilitate collaborative study; foster inter-disciplinary discourse; promote experimental learning; and develop a more theoretically nuanced and historically grounded starting point for critiquing transparency and its operations within digital culture.

Convenor: Tom Kissock (CDH Methods Fellow)

This Methods Workshop will offer Video Data Analysis for Social Science and Humanities students. It’s a relatively new, broad, and innovative multi-disciplinary methodology that helps students understand how video fits into modern research both inside and outside academia. For example, Cisco has estimated that video will make up 80% of internet traffic and 17.1% of it will be live video which is a 15-fold increase since 2017; therefore, it’s a tool that cannot be overlooked when conducting research.

Tom will address how to use video ethically, for example:

• Informed consent
• Storage
• Privacy

and also practically;

• Building timelines
• Coding schemes
• Presenting research findings

Tom will also plans to include a lesson focussed on viewing livestreams in a reflexive manner as this is a huge topic in the TikTok era

About the convenor: Tom has fifteen years’ experience as a Director, Executive Producer, and Livestream expert for the BBC, YouTube, NBC, and Cisco; coupled with seven years’ experience researching video witnessing and human rights abuses. In 2020 he received his MSc in Globalization and Latin American Development from UCL where his research used Video Data Analysis as a research methodology. He tracked how populist politicians in Brazil built misinformation campaigns by strategically cross-sharing videos to avoid journalistic questioning as a symbolic accountability mechanism during the 2018 presidential elections.

His PhD in Sociology at the University of Cambridge is a loose extension of his MSc, but explores positive aspects of streaming advocacy, such as how Indigenous video activists in Brazil use live video on platforms like Instagram, TikTok, and Kwai to reach audiences to discuss climate change, the environment, and land rights. He is interested in how video can produce knowledge and, subsequently how societies value different knowledge through the process of video witnessing. In his spare time, he serves as the Executive Producer of Declarations: Human Rights Podcast (part of Cambridge’s Centre for Governance and Human Rights), has given lectures on live streaming and human rights at MIT, UCL, and the University of Essex, and has written pieces for LatAM Dialogue and the Latin American Bureau.

Convenor: Dr Eleanor Dare (CDH Methods Fellow)

This Methods Workshop will invite participants to originate innovative research methods using virtual and augmented reality technologies underpinned by theoretical and pedagogic understandings. The session is conceived in recognition of an increasing interest in using virtual and extended reality (VR and XR) to create collaborative research spaces that span different locations, time zones, and spatiality. Such spaces might be used to investigate the impact of design, architecture and location on education or new ways to teach an array of subjects, from language to mathematics to performance, AI ethics and music.

About the convenor: Eleanor is currently the Co-Convenor for Arts, Creativity and Education at the University of Cambridge, Faculty of Education, they are also the Senior Teaching Associate: Educational Technologies, Arts and Creativity, lecturing and supervising on MPHIL Arts, Creativities and Education, MPhil Knowledge, Power and Politics, and MEd Transforming Practice. Eleanor is module lead for AI and Education, a Personal and Professional Development course at Cambridge.

Eleanor Dare’s research addresses the implications of digital technology and virtuality as a material for collaboration, critical-educational games development, performance, worldbuilding and pedagogic experimentation. Eleanor has been involved in several AHRC/EPSRC/ESRC/Arts Council/British Council funded projects investigating aspects of virtual and extended reality as well as projects with the Mozilla Foundation (AI-Musement/Monstrous 2022-2023), Theatre in the Mill Bradford (Bussing Out, 2022) and the Big Telly Theatre Company (via the Arts Council of Northern Ireland) for Rear Windows, forthcoming.

Dr Anne Alexander, Cambridge Digital Humanities

Places are limited and participants must complete this form in order to participate in addition to booking online. We will write and confirm your participation by email. Bookings will remain open until 10am, 11 October 2021; However, participants are encouraged to apply early as demand is likely to be high.

This online workshop will provide an accessible, non-technical introduction to Machine Learning systems, aimed primarily at graduate students and researchers in the humanities, arts and social sciences. It is designed as a preparatory session for potential applicants to our Interaction with Machine Learning Guided Project which will run in Lent Term 2022 in collaboration with the Department of Computer Science and Technology. However, it can also be booked as a standalone session.

Interactive Fiction (IF) stories let readers decide which paths the story should follow, featuring non-linear narrative design. The discipline combines the excitement of post-structuralist narratives with the power of creative coding, making it a perfect introduction for participants more familiar with one field than the other. In this workshop, led by Methods Fellow Claire Carroll, we’ll explore both parser-based (rooted in reader instructions and terminal response) and choice-based (hyperlink or multiple choice-driven) IF and work together to write our own interactive fiction. The workshop will also introduce participants to the passionate IF community, which offers advice and support to experienced writers and newcomers alike.

This CDH Basics session explores how data which you have captured rather than created yourself, is likely to need cleaning up before you can use it effectively. This short session will introduce you to the basic principles of creating structured datasets and walk through some case studies in data cleaning with OpenRefine, a powerful open source tool for working with messy data.