COMPASSEstimating indicators of human development in South Africa from satellite images using machine learning
Emily-Rose, Martin Bekker, Ken J. Nixon
This research investigates the extent to which socioeconomic indicators that reflect human development can be estimated from satellite images. The investigation extends previous work that incorporates nighttime lights into a deep-learning model to estimate asset wealth and consumption expenditure. Here, we apply this approach to estimate access to infrastructure and basic services in a South African context. This research is a stepping stone in the direction of estimating a variety of indicators from low-cost satellite data regularly. This would provide critical complementary data to current development surveys, especially in data-scarce countries.
Data Driven Remote Sensing Approach For Peri-Urban Demarcation of Hyderabad City, India
Soumil Hooda, Ravi Bhushan, Hiten Vidhani, Manik Gupta, Lavanya Suresh, Timothy Clune
With rapid urbanisation, the centres of urban transformation across the metropolitans in India have shifted to their peripheries, leading to the creation of peri-urban areas. These areas with both urban and rural attributes are undergoing dynamic socio-spatial and physical changes. Despite their importance in urban transformation, there have been limited administrative efforts to demarcate these areas which leads to improper land zoning and inefficient governance. There is a lack of existing studies from the global south that have tried to understand the peri-urban expansion spatially and temporally and demarcate peri-urban boundary in the last decade. This paper uses spatial parameters to propose a data driven remote sensing approach for demarcating peri-urban areas in Hyderabad Municipal Region (HMR) in India. The model using a combination of thresholding and Support Vector Machine (SVM) has been used to demarcate peri-urban areas and it has been demonstrated that peri urban area has increased by 95.1 per cent between 2013 and 2020. The proposed method is significant for developing countries where the timely availability and quality of socio-economic data is a challenge.
Milk Matters 4.0: Challenges in Deploying University-Led Mobile Application Development for Small NGOs
Deborah Talbot, Melissa Densmore
Milk Matters is a small Cape Town based non-profit milk bank. Their primary role is to collect expressed breastmilk from donor mothers, pasteurize it and distribute it to recipient infants in need. Previous postgraduate projects from the University of Cape Town (UCT) have co-designed a donor facing mobile application with Milk Matters, however no mobile application is currently deployed or promoted by the non-governmental organization (NGO). This project will build upon the work already done with Milk Matters and aims to update the full system for deployment. While post-deployment evaluation will also analyse the uptake and usage of the application, this poster will focus on discussing the challenges in the deployment of university-led mobile application development for small NGOs.
AI-Based Platform For Predicting The Risk of Having NCDs
Ariane Shimirwa, Claudine Mahoro, Tyson Muvunyi
Non-communicable diseases (NCDs) are chronic medical conditions not caused by infectious agents and primarily associated with lifestyle factors. This paper highlights the significant impact of non-communicable diseases (NCDs) on global public health, particularly in low- and middle-income countries where they account for most premature deaths. A privacy-preserving AI platform is proposed as a solution to predict the risks of having NCDs, such as cardiovascular diseases and type 2 diabetes, using clinical data and advanced machine learning algorithms to reduce the number of deaths that occur yearly. The platform alerts individuals with high risks and provides them with further medical consultation options.
Mapping Construction Grade Sand: Stepping Stones Towards Sustainable Development
Ando Shah, Suraj R Nair
Sand and gravel are critical inputs to economic growth as the primary constituents of concrete and asphalt. While demand for these materials has skyrocketed due to large construction and reclamation demands, rates of extraction are unsustainable and result in adverse environmental and socio-economic consequences, especially in the Global South. Excessive sand and gravel mining threatens biodiversity and hydrological functions, heightens the risk of damage to critical infrastructure, and increases vulnerability to extreme climatic events. In this poster, we argue that mapping the world’s sand and gravel resources is the first step towards informing effective policy that can ameliorate these harms while achieving sustainable development. We have developed flexible machine learning algorithms which can detect usable sand and gravel resources in river basins and coastlines at global scale with high spatial resolution (10 m). Our approach uses object based image analysis methods fusing freely available Sentinel-1 and Sentinel-2 multispectral satellite datasets. This method achieves an F1 score of 78.53% and accuracy of 79.59% using a random forest classifier trained on a global dataset of in-situ grain size observations. We further validate performance in sections of the River Ganga where a gravel to sand transition is known to occur, and in a section of the River Sone where a number of known sand mine concessions exist. This work lays the foundation to build end-to-end deep learning models that can predict where illegal sand mining occurs.
Examining the Features of Mobile Apps for Environmental Sustainability
Sarah Cooney, Jeremiah Matthew, Ava Ferrentino
This short paper presents initial results from an ongoing effort to collect data on commercially available applications for promoting environmental sustainability. It discusses the preliminary searches and the resulting set of 76 apps. The paper details initial insights from analyzing the features of these apps and describes areas for future work.
Designing Empathic VUIs: A study of Non-Verbal Vocal Cues of Synthetic Speech
Riya Singh, Anupriya Tuli
The emotional value embodied in audio responses of voice user interfaces (VUIs) impacts how users interpret, experience, and understand these interfaces. The emotion delivered in a voice response depends both on its verbal content and non-verbal vocal cues. We undertook a preliminary investigation where we used non-verbal vocal cues of pitch, speech rate, and intonation to modulate the emotion embodied in a synthetic speech to generate an empathic voice response. Using the Wizard of Oz, we investigate how matching Alexa’s response emotion with the user’s affective state impacts the overall user experience. Our results establish the potential of non-verbal vocal cues of a synthetic speech as crucial parameters to generate empathic voice-only responses. We also noted that our approach improves the likability and usability experience with VUIs.
CoLRN – A Community-Based Vision for Local Resilient Networks
Ndinelao Iitumba, Siddhant Shinde, Deysi Ortega, Naveen Bagalkot, Nervo Verdezoto, Melissa Densmore, TB Dinesh
In this research, we share our findings from a series of design workshops with community wireless network members and their users in India and Africa to develop a community-based vision for resilient local networks. We simultaneously leveraged existing projects in India and South Africa around network management interfaces and local content creation to evaluate our design strategies to foster resilience and effectiveness in empowering community networks. Through this work, we identified the challenges and opportunities for innovative approaches to leveraging networked technologies to bring communities together to learn from each other on how they manage and use their community network. We highlight key opportunities to explore a) infrastructural resilience through community-centred design of network management tools, and b) novel approaches to support content creation tapping community desires to capture local knowledge, through annotation of digital stories and production of radio content.
An Educational Ecosystem Based on Blockchain
Rudaiba Adnin, Rezwana Reaz
Blockchain as a new technology plays a significant role in education, yet, there exists a limited number of research on blockchain-based educational ecosystems. Therefore, in this paper, we propose a novel educational ecosystem integrating four stakeholders (students, instructors, institutions, and employers) for a wide range of academic functions such as cross-institutional study, skill development, academic assessments, certification, and employment opportunities in a blended learning environment. Moreover, the proposed user-centric system tackles issues regarding potential bias and tampering with the educational records of students. We further discuss the advantages, challenges, and possibilities that arise within this proposed ecosystem.
Exploring Deployment and Adoption of Locally hosted Digital Services with Communities
Lizalise Luxande, Holly Judge, Shreeya Khoosal, Melissa Densmore
This paper discusses the implementation and development of 3 ser- vices on the iNethi community wireless network platform, namely – a community radio, parenting chatbot and community exchange platform. The implementation timeline, design considerations and project aims are scaffolded within this proposal with the overarch- ing objective of empowering individuals in the Ocean View com- munity through democratizing access to local services by hosting them on the iNethi network and making them “zero-rated”.
User-Agent Interactions in Mobile Money Banking in Kenya and Tanzania
Karen Sowon, Edith Luhanga, Lorrie Faith Cranor, Giulia Fanti, Conrad Tucker, Assane Gueye
Digital financial services have catalyzed financial inclusion in Africa. Commonly implemented as a mobile wallet service referred to as mobile money (MoMo), the technology provides enormous benefits to its users. While the benefits of mobile money services have largely been documented, the challenges that arise especially in the interactions between human stakeholders remain relatively unexplored. In this study, we investigate the practices of mobile money users in their interactions with mobile money agents. We conduct 72 structured interviews in Kenya and Tanzania (n=36 per country). The results show that users and agents design several workarounds in response to limitations and challenges that users face within the ecosystem. These include advances or loans from agents, relying on the user-agent relationships in place of legal identification requirements, and altering the intended transaction execution to improve convenience. The results suggest a need for rethinking among other things the privacy, security and usability components of the ecosystem, as well as policy and regulatory controls to safeguard interactions while using mobile money.
Scaling Carbon Footprinting: Challenges and Opportunities
Bharathan Balaji, Geoff Guest, Gargeya Vunnava, Jared Kramer, Aravind Srinivasan, Michael Taptich
Rapid and continuous increase in greenhouse gas (GHG) emissions is warming our planet at unprecedented rates. Consumer products and services, including all aspects of the corresponding supply chain, contribute to more than 75% of these emissions. Attribution of GHG emissions to each product will drive awareness and change from individual consumers to large corporations that produce and own these products. However, accurate and standards-compliant accounting of carbon emissions for millions of products is challenging as it requires detailed manufacturing and supply chain data, and subject expertise in life cycle assessment (LCA). We posit that ideas from computer science and machine learning can alleviate bottlenecks in LCA, and research contributions from this community will accelerate solutions for accurate carbon footprint estimation as well as carbon abatement strategies at scale. We present the principal components of an LCA study with a step-by-step walk-through. We elaborate on the challenges to scale LCA, and identify the opportunities to innovate in this space with techniques such as information extraction, personalized recommendations, and decision making under uncertainty.
Inclusion Drives Sustainable Development: The Case of Social Robotics for Africa
Pamely Zantou, David Vernon
The achievement of sustainable development goals requires collaboration among all stakeholders. In turn, this necessitates that the environment in which solutions are developed be inclusive of all. Artificial intelligence (AI) is widely recognized to be a powerful enabling technology that can be used to leverage these solutions. However, AI, including the increasingly important field of robotics, is not inherently inclusive and much remains to be done in the democratization of AI. In this paper, we argue that inclusivity cannot be achieved without cultural sensitivity being factored into the design of AI and robotics technologies. This papers presents culturally sensitive human-robot interaction in social robotics as one example of endeavours to achieve this inclusivity and, thereby, drive sustainable development in Africa.
Power Analysis of a Large-Scale Solar-Powered Urban Sensor Network
Alex Cabral, Jim Waldo
Solar power is often touted as a reliable renewable energy source for low-cost sensor networks in various environments. However, there have not been extensive real-world studies to examine how well solar-powered sensor networks perform in urban settings over long periods. In this work we analyze the performance of a large-scale solar-powered sensor network over one year in Chicago, Illinois. We find that over 35\% of the devices experienced charging issues between the months of October and March, resulting in over 33,000 hours of data loss. Surprisingly the devices that had issues charging were not all located near tall buildings and were often found in majority Black and Latine neighborhoods. These findings highlight the need for continued research in alternative power sources and energy harvesting techniques, and increased real-world deployments to identify additional barriers in using sensor networks for real-time monitoring in cities.
Systematic analysis of the effectiveness of adding human mobility data to COVID-19 case prediction linear models
Saad Mohammad Abrar, Naman Awasthi, Daniel Smolyak, Vanessa Frias-Martinez
Human mobility data has been extensively used in COVID-19 case prediction models. Nevertheless, related work has questioned whether mobility data really helps that much. In this paper, we present a systematic analysis across mobility datasets and prediction lookaheads and reveal that adding mobility data to predictive models improves model performance only for about two months at the onset of the testing period, and that performance improvements – measured as predicted vs. actual correlation improvement over non-mobility baselines – are at most 0.3.
Why SuaCode?”: Understanding African Students’ Motivations for Taking a Smartphone-Based Online Coding Course
Michael Addo, Nana Maryam, Victor Kumbol, Judith Uchidiuno, George Boateng
Computer programming MOOCs are instrumental in providing students with high-quality instruction in areas where there is limited access. They are especially beneficial to post-secondary African students as less than 1% of them leave secondary school with fundamental coding skills. One strategy for increasing their efficacy for African students is to understand students’ motivation for enrolling. These insights can inform the design of MOOC content and assessments to align with students’ interests. We administered an open-ended response survey to (self-identified) Africans enrolled in a smartphone-based online coding course (SuaCode). We analyzed a random sample of 450 (of 3000) responses using a grounded theory approach. We found that most African students (68.7%) participated in SuaCode for intrinsic reasons such as improving themselves, learning with like-minded individuals, and gaining skills to help address societal issues. We discuss the implications of these findings in the design of programming MOOCs targeted at African students.
Understanding Black People Building Technology For Black Lived Experiences
Lisa Egede, Leslie Coney, Brittany Johnson, Christina Harrington, Denae Ford
The HCI field has seen a heightened interest in understanding how racially minoritized people create and foster community across various platforms. While the growth in this research space is evident, little work has been done around understanding how Black creators build and design technological systems for the needs of their own communities. In this poster we present findings from our study conducted with Black technologists from a wide array of domains, with the goal of highlighting their experiences using, creating or curating resources to support the Black Lived experience. Concluding this work we found that technologists take a multifaceted approach to design as a means of survival, to stay connected, for cultural significance and to celebrate Black joy. Finally, we discuss how designing for the Black experience extends beyond tackling inequities and that taking various approaches to supporting technologists building for their own communities can lead to more impactful outcomes.
A Mean Field Game Approach to Promote Sustainable, Socially Optimal Behavior in
Rational Individuals for Effective Management of Epidemics
Amal Roy, Pranoy Das, Chandramani Singh, Soumyarup Sadukhan, Yadati Narahari
When an epidemic strikes, it becomes crucial to effectively contain and suppress its spread in order to minimize the loss of lives and alleviate the burden on the public healthcare system. Numerous non-pharmaceutical and pharmaceutical interventions have been extensively explored to tackle and limit the spread of epidemics. However, it has been observed that despite the imminent threat posed by an epidemic, individuals often exhibit rational behavior and exercise their free will instead of adhering to best practices tailored for epidemic control. This, unfortunately, leads to potentially undesirable consequences. In this paper, we study an appropriately formulated epidemic game that involves rational individuals, aiming to identify strategies that can induce socially optimal behavior. By employing a mean field approach, we derive a centralized control policy that optimizes societal well-being, as well as a Nash equilibrium-based control policy. We conduct carefully designed thought experiments which highlight various policy measures and refinements in non-pharmaceutical interventions, to promote sustainable, socially optimal behavior in rational individuals. Notably, our study is conducted within the realistic context of vaccine availability, further enhancing its practical relevance.