top of page

Abstract:

The objective of this project is to develop AI Frameworks based on decolonizing data, algorithms, and visualization in the Global South. We learn from local intelligence, which includes faith-based beliefs, myth, language, and arts and crafts. In our ongoing study of the knowledge, materials, and politics in rural Bangladeshi local art, culture, visuals, and computing in rural witchcraft, Nakshi-Katha, and Hindu idol-making practices, we developed an understanding of culturally appropriate data practices that help us develop decolonial AI-framework. 

We employ this knowledge in engaging and codesigning culturally appropriate AI-mediated agricultural infographic tools with rural farmers in Bangladesh. Like locals of many other parts of the Global South, Bangladeshi farmers have long-practiced indigenous visual communication techniques. In our ongoing design project, we address the cultural mismatch between modern scientific design principles in AI and traditional values in the Global South, which affects the usability of many AI and data-driven systems. Particularly, agricultural advice on weather, climate, soil quality, harvest, and among others, when conveyed through conventional digital ways of information communication, fails to communicate with many rural farmers, as our long-term research in rural Bangladesh informs us.

Project Goals:

Broader goal: To develop a culturally appropriate AI framework to address environmental justice concerns in the Global South. 

Specific goal: To co-design AI-mediated agricultural infographic tools with rural Bangladeshi farmers based on their long-practiced indigenous techniques and rural Bangladeshi narrative visualization and data curation methods.

Decoloniality-data-algorithm-visualization.png

A witch in Kazipur village showed several example Jantras used for Tantra in their witchcraft practice. (Clockwise from left) Jantra for solving marriage issues, fo witctcraft training, for dismissing effects of bad spells, and for better health.

katha_22_edited.jpg

A rural women from Mandartola village showed us her great-grandmother’s Nakshi-Katha explained that the wheel from the top middle of the main Nakshi-Katha what has the monetary records of 16 of the productive lunar cycles (8 months) that year,

Background:

Many agriculture-based Global South countries’ development policies associated with environment and climate change mismatch with their traditional farming practices. In Bangladesh, for example, rural farmers are disconnected from national and international guidelines regarding environment preservation due to their lack of understanding of modern digital data representation and narrative construction methods (e.g., tables, graphs, charts, etc.). Like many other countries, Bangladesh faces degradation in soil quality, low harvest, and rapid decline of water level — resulting in an unstable local food market and, hence, the country’s economy.

We develop an AI framework to address environmental justice in the Global South based on local indigenous practices of data curation and visual narratives. We integrate rural Bangladeshi visual communication methods to develop AI-mediated agricultural infographic tools so that rural low-resource and low-literate farmers can engage with environmental justice discourses and make informed decisions. We build on decolonial AI literature and pluralistic co-design for environmental justice.

 

Initial Fieldwork and Findings:

We engage with rural artists and farmers in Jessore, Bangladesh, using ethnographic techniques. We employ participatory observations, contextual inquiries, storytelling, interviews, and focus-group discussions in this regard. Additionally, we frequently engage with them in codesign sessions to brainstorm ideas. 

Our Initial findings bring in knowledge of local Bangladeshi intelligence of data curation, visual narratives, and infographics from rural arts and crafts, including visual witchcraft, Nakshi Katha, and Hindu Idol making. In an ongoing work, we are engaging with rural Bangladeshi farmers on their daily-life information queries and use of digital platforms and their interpretability. We use this understanding of their data literacy in designing culturally appropriate AI-infused infographics.  

image.png
image.png

A rural Bangladeshi farmer is plowing his field while being interviewed with the ethnographer.

A group of rural Bangladeshi male and female farmers participating in Focus Group Discussion with the ethnogrpaher.

Selected publications and activities: 

  • [CHI 2023Abstraction and semiology in constructing visual narratives,

  • [FAccT 2022] Big data and AI in Global South [Workshop], 

  • [Mozilla Festival 2022] Decolonizing AI [Workshop],

  • [ECSCW 2022] Fieldwork reflections on pictorial consent,

  • [CHI 2021] Understanding luck, hunch, faith, and data-driven prediction in betting

  • [CSCW 2021] Rural fact-checking procedure for misinformation [🏅Diversity and inclusion recognition]

  • [CSCW 2020] Contrasting grammar of modern data visualization and local traditional visualization

  • [ACM Interactions 2020] Rural faith-based practice in combating COVID-19,

  • [CHI 2019] Witchcraft and HCI: Morality, modernity, and postcolonial computing.

Project Team:

Our core team includes international and Bangladeshi researchers, local partner NGO Rural Reconstruction Foundation (RRF), and our existing field connections in rural villages in Bangladesh. 

Sharifa Sultana is an Assistant Professor of Computer Science at the University of Illinois Urbana-Champaign, USA. She is a human-computer interaction (HCI) designer and critical computing researcher. She engages with justice concerns, including designing decolonial data-driven systems and AI, computing for alternative rationalities and moralities, and gender justice in computing. She deploys a variety of qualitative, quantitative, and design methodologies to probe and address social justice agendas in low-resource, marginalized communities. She has more than 7+ years of experience conducting ethnography, co-design, and deployment with rural villagers in Jessore, Bangladesh. She builds and evaluates computing technologies, including accessible, low-cost, and intelligent mobile and web applications to improve the quality of their life.

Jeffrey M. Rzeszotarski is an Assistant Professor in the School of Information Science at Cornell University, USA, with interests in data visualization, crowdsourcing, and social computing. His research focuses on helping both experts and everyday people make sense of data. His research aims to make big and small data accessible to as many people as possible, and he endeavors to encode that value into the systems he designs. Together with Sultana and Ahmed, he has been investigating rural Bangladeshi traditional and contextual visual communication, data curation, and visual narrative-making practices in local arts, crafts, and performances for the past four years to develop decolonial data-driven frameworks. 

Syed Ishtiaque Ahmed is an Assistant Professor of Computer Science at the University of Toronto, Canada. He directs the Third Space research group at the DGP Lab. He is a graduate faculty member of the School of Environment, a Faculty Fellow at the Schwartz Reisman Institute for Technology and Society, and a Senior Fellow at Massey College. His work is motivated by social justice and sustainability issues, and he puts them in the academic contexts of HCI and Information and Communication Technology and Development (ICTD). His technical and methodological apparatuses include ethnography, design, NLP, and tangible UI. He has 12+ years of experience designing, developing, and evaluating computing technologies with marginalized communities. 

 

Salim Reza is the Director of the Microfinance Program at the Rural Reconstruction Foundation (RRF), Bangladesh. He has been working with RRF for 17+ years. He holds a Master’s degree in Accounting. Some of the notable World Bank-supported projects led by Reza include the Sustainable Enterprise Project on flower cultivation (SEP), the Low Income Community Housing Support (LICHS) Project, and the Recovery and Advancement of Informal Sector Employment (RAISE). With RRF, he has grown extensive experience in agriculture financing projects, particularly in agriculture production and agriculture machinery; some of his led notable ongoing projects with RRF are described in the attachment. He directs RRF’s operation in 34 districts (among 64 districts) in Bangladesh. 

Support team in the field: 

In addition to RRF’s support team, we have local Bangladeshi software developers and field researchers affiliated with local Bangladeshi industry and research institutions who have been conducting research with us in partnership for many years. Their contribution will be sought in different stages of this project. 

Contributors:

Codesign sessions will involve local farmers, rural artists, practitioners, seniors, AI and scientific data visualization experts, and RRF fieldworkers to sit together and develop a framework of AI-mediated infographics that they can use to communicate with each other. We will aim to develop a value-sensitive visualization grammar that will uphold the traditions and faiths attached to the local community, local materials, and local history.

image.png

University of Illinois Urbana-Champaign, USA

image.png

Cornell University,

USA

image.png
image.png

University of Toronto,

Canada

Rural Reconstruction Foundation, 

Bangladesh

AI and Environmental Justice in the Global South

paper clip_edited_edited_edited.png
bottom of page