A round table discussion with Saa Dittoh, Julia Neidhardt, Marcel Stelte, Auke Schuringa, Oleksandr Zakotianskyi, took place at VU Amsterdam, this was a session of the Conference “Navigating the Anthropocene”, Amsterdam, 30 January 2025. During this event, two major frameworks for a better world are linked. These are (1) the UN’s Sustainable Development Goals, c.q. SDG1 zero hunger, SDG5 gender equality, and (2) Digital Humanism, the initiative that aims to ensure that digital technology remains centered on human interests. Three recent EURIDICE master research projects are discssed.
Prof Saa Dittoh (UDS, Tamale, Ghana): Integrating Indigenous and Scientific Knowledge for Sustainable Food Systems in Africa — The Plug-In Principle.
Abstract In academic research, a negative bias towards traditional knowledge systems is often present. This ignores the importance, experiences, and potential value of indigenous knowledge. The plug-in principle is a theoretical and practically oriented framework, proposed as an alternative to the dominant narratives of how to ensure appropriate and effective implementation of development interventions and the adoption of appropriate technologies to food production. It is a guiding a new approach towards integration of knowledge systems. The perspective of the African smallholder farmer, who represents the largest producer of food on the continent, is central. The plug-in principle starts from existing situations and knowledges and tries to “better” the situations by plugging-in new ideas, techniques and practices. Saa Dittoh is the editor-in-chief of a new Open Access book with this title, scheduled to appear in 2025 at Springer Nature, Heidelberg, in its SDG series (SDG-2, zero hunger).
Dr Julia Neidhardt (TU Wien, Vienna, Austria): SDGs and Digital Transformation in the Light of Digital Humanism.
Abstract The Digital Humanism Initiative is an international collaboration seeking to build a community of scholars, policy makers, and industrial players who are focused on ensuring that technology development remains centered on human interests. Technology profoundly shapes the world we live in, and the stakes are high. Digital Humanism deals with digital technology development and policies based on human rights, democracy, inclusion, and diversity. Among others, the Digital Humanism initiative has published two Open Access books with many studies of the issues involved and a textbook (Springer 2022 and 2024), that together now have more than 1.1 Million downloads. Also, a book for use by high schools and their pupils has been written (2024) that is being translated in several languages. Julia Neidhardt is assistent-professor at the Centre for AI and Machine Learning of TU Wien, Austria, and UNESCO co-Chair in Digital Humanism.
Moderator of keynote talks: Hans Akkermans
Saa Dittoh is professor of agro-economics at the University for Development Studies UDS, Tamale, Ghana.
Julia Neidhardt is assistant professor at TU Wien’s Centre for AI and Machine Learning (CAIML), and UNESCO co-Chair for Digital Humanism.
11:20 – 12:00 Round-Table including the keynote speakers and three student ICT4D – Digital Society & Global Citizenship master project pitches & posters.
Moderator of Round Table: Anna Bon
Marcel Stelte (VU Computer Science alumnus 2024) – Gender Bias against LGBT+ in Software Design
Abstract. Bias towards otherness is a reality in platforms or applications, and can be discriminatory and often detrimental towards minority groups, for example in gender. Is this bias introduced already by design, as part of the software engineering methodologies? And can this bias be explored, mitigated and inclusiveness increased? This is the problem tackled in the master research project by Marcel Stelte, focusing on bias in software design, towards members from the LGBT+ communities, who are often under-represented in many aspects of Society. This study is aligned with SDG5: Gender Inclusiveness.
Oleksandr Zakotianskyi (VU Computer Science student)– Predicting War with AI — Explainable Probabilistic Forecasting of Conflict-Related Fatalities
Abstract. Can we predict war using data and intelligent models? And how certain would we be in our predictions? Policymakers are interested in uncovering the conflict dynamics and making robust forecasts to take anticipatory actions and reduce the impact on vulnerable people. At the same time, regular citizens are looking for accurate and not biased forecasts for their own security and awareness. The master project by Oleksandr Zakotianskyi provides the first publicly available and explainable early conflict forecasting model, capable of forecasting the distribution of conflict-related fatalities on a country-month level. [SDG-16]
Auke Schuringa (VU & UvA Computer Science & AI student)– Energy-efficient Speech Matching of Indigenous Languages with AI
Abstract. For rare indigenous languages, of which there are many in the Global South, speech applications are still largely lacking, due to high costs and high energy consumption to create these models. Auke Schuringa proposes an alternative method that requires less data and is more energy efficient. It uses an already trained model for a phonetically close popular language to create a phonetic transcription of a recording, thus generalizing context-specific or unique language specific aspects out to create a broader phonetic description of the spoken segments, and calculating similarities between phonetic descriptions to allow for matching spoken words. The obtained results open possibilities for creating speech-based applications. This method is energy-efficient and is therefore much more environmentally sustainable than conventional machine learning solutions that are commonly used to resource languages. [SDG-8/9]
W4RA Community – The Web alliance for Regreening in Africa
Abstract. W4RA is a trans-disciplinary community for research and action on real world topics related to the Digital Society. It brings together academics and non-academics with a variety of backgrounds and knowledge domains. This includes information science, computer science, artificial intelligence, organization, business, and communication sciences, sustainable land management, rural development and web science, but also indigenous knowledge. W4Ra is hosted as a research and educational program at the Vrije Universiteit in Amsterdam. Since its inception in 2009, W4RA has been inspired by the African Regreening Initiatives, initiated and implemented by local farmers in Africa.
This was an EURIDICE public dissemination event.