February 2, 2019 QCAIUser

AI for Social Good – Abstract + Slides

#NextGenGov: Reimagining Governance Models
UNDP will present its perspective on the use of data, and data-centric technology, to reimagine governance models and mechanisms that are needed to better address the challenges of the 21st century and accelerate progress towards the achievement of the Sustainable Development Goals. We will discuss our exploration of data as an enabler of and catalyst for increased accountability, transparency and trust; more targeted service delivery; more accurate and timely measurement; among other things. We will present some emerging cases and areas of exploration, such as the future of cities, digital economy, migration and prevention of violent extremism, for all of which UNDP welcomes data partnerships across sectors.
AI for the Most Vulnerable

Digital technologies have the potential to create opportunities to accelerate sustainable development. However, at the moment, there are deep inequalities in access to technology and digital products, the lack of which isolates vulnerable communities from opportunity. At the Office of Innovation we work at applying emerging technologies, such as AI, ML, drones, VR/AR, and distributed ledgers for the benefit of children worldwide. We partner with private companies, humanitarian organization, and governments to help us build a better, more equitable, and sustainable world.

Data Innovation at UNHCR
The presentation goal is sharing-knowledge about the different data-related projects currently being undertaken by UNHCR Innovation service. In addition, to portray the ongoing research by some of our service partners particularly focusing on forced displacement. Furthermore sharing some of our way of developing projects: field-oriented, focusing on innovation as a process for cultural organizational change with particular emphasis in data science techniques and artificial intelligence tools.
Using AI to Detect, Map and Quantify Internal Displacement
Disasters and conflict displace tens of millions of people every year, affecting nearly every country around the world. More timely, accurate information about the number and location of people displaced — or at risk of imminent displacement — can save lives and protect livelihoods. Using natural language processing, machine learning and other techniques, IDMC has developed tools to analyze non-traditional sources of data in order to estimate displacement risk, identify incidents of internal displacement and monitor these situations as they evolve over time.
AI for Accelerating Sustainable Development & Humanitarian Action in the Asia Pacific

Big data presents new opportunities for governments and development organisations to access timely insights that can inform evidence-based decision and policy making. As a data innovation lab, Pulse Lab Jakarta is working to close information gaps in the development and humanitarian sectors through the adoption of big data, real-time analytics and artificial intelligence. From advanced data analysis of mobile network data, social media data, and other emerging new sources of data, this presentation highlights the Lab’s work across the Indonesian archipelago and the Asia Pacific in partnership with government institutions, private sector players and civil society.

Gender Gaps in Urban Mobility
The availability of large-scale digital data on human mobility has fundamentally changed the way we study cities and urban mobility. Research so far has concentrated on aggregated data, thus neglecting important potential biases such as gender. Here we analyze gender-disaggregated Call Detail Records and present a gendered view on mobility in a major Latin American city. We characterize gender differences in the diversity and types of visited locations, study these differences in relation to other socio-economic features, and discuss insights and hypotheses about factors driving the observed inequalities.
The Cross-field Applications of Data Science for Social Good
At Vodafone Research we have a line of work devoted to Data Science for Social Good, where we leverage our global footprint to perform research with humanitarian purposes. With a global reach of over 70 countries and more than 313 million customers, and through its multiple partners, Vodafone is very well placed to be able to do Social Good research projects. A member of the Digital Impact Alliance, GSMA’s Big Data for Social Good and in close collaboration with the United Nations and several NGOs, we have been working on Public Health, Financial Inclusion, Humanitarian Crisis Management, Social Capital and Population Studies. In my talk, I will give an overview of the Data Science Research projects in the area of Social Good, I will briefly present two examples --one on public health and the other one on financial inclusion-- and I will share our view on the existing challenges to realize our vision of a better world thanks to the existence of large-scale human behavioral data.
Understanding migration flows with mobile data: Challenges & Opportunities
By 2050 as many as 17 million people in Latin America could be forced out of their homes by persistent slow onset climate change. This is particularly the case in low and middle-income countries, where IDPs are more likely to live in remote locations with poor infrastructure, or may be surrounded by volatile security situations. Thus, measuring migration patterns becomes a very complex task, compounding the economic and social impact on these individuals, families and communities. In Latin America, Telefonica and the United Nations Food and Agriculture Organisation (FAO) have partnered to address this knowledge gap through mobile Big Data. We investigated the case of IDPs in Colombia likely motivated by extreme climate variability. By using mobility patterns as a proxy for human behavior, Telefonica was able to measure and map previously unseen migration patterns. The result of the study is promising in terms of the potential value of telco data for the study of the migration phenomena. Nevertheless, there are some challenges remaining that would need to be addressed in order to unlock its full potential.
The Data for Refugees Challenge: Sharing mobile CDR data for improving the conditions of Syrian refugees in Turkey
The Data for Refugees (D4R) Challenge was a non-profit challenge initiated to improve the conditions of the Syrian refugees in Turkey by providing a special database to the scientific community for enabling research on urgent problems concerning refugees, including health, education, unemployment, safety, and social integration. Collected from 1 million customers over a one year period, the database shows activity and movement of refugees and citizens over the entire country.
AI 4 Good – An Overview of IBM Research Science for Social Good Program
In this talk I will describe our Science for Social Good program where we have identified important real-world problems in collaboration with partners and developed artificial intelligence-based solutions to address their problems. This program allows social good fellows (post doctoral researchers and interns) to work exclusively on these exciting problems during their stay at IBM Research. I will also share some key insights we learned in this process over the past two years of working with a wide range of organizations.
Data Collaboratives: The Emergence of Public-Private Partnerships around Data for Social Good
In the past few years, The GovLab has sought to understand pathways to make policymaking and problem solving more evidence-based and data-driven. Our work is driven by a recognition of the potential of use of privately processed data through Data Collaboratives — a new form of public-private partnership in which government, private industry, and civil society work together to release previously siloed data, making it available to address the challenges of our era. Our research suggests that Data Collaboratives offer tremendous potential when implemented strategically under the appropriate policy and ethical frameworks. Nonetheless, this remains a nascent field, and several barriers limit the systemic use of Data Collaboratives.
Use of Facebook advertising data for humanitarian action
An overview of the work between iMMAP and QCRI in using Facebook ad data to calculate the population distribution and growth trends of Venezuelan migrants in Colombia and the wider region. A description of the methodology, its quality, and its final use with an audience that includes humanitarian decision-makers in the Latin America region. Future plans for potential application of the results in other emergencies and in the wider region.
Leveraging Big Data, Open Algorithms and AI for Human Development
Do Big Data, algorithms and AI threaten human development and democracy, or do they offer the possibility of building future human systems where decisions will be more rational, policies more efficient, processes more inclusive, politicians more accountable? Can we think and work constructively beyond a naïve embrace or systematic fear mongering of all things AI? In particular, how can societies and communities “leverage” AI by applying both its key tools and principles to concrete projects and policies to build an AI that reflect and serve human development?
Data Aurora in the Mena Region: achievements, opportunities, and challenges
The maturity of the data innovation landscape has increased enormously since 2012. Back then, our “data discussion” were theoretical. Today, after 6 years of working in different industries, (media, retail, tourism, etc...) along with many partners (Governments, NGOs, Academic institutions, etc...), we realize that we have established a large skillset in the data science and machine learning domain. In this presentation, we would like to share the major data and social network analytics techniques we have developed while approaching challenges in various domains. Also, we would like highlight the challenging scenarios one can face while working with data in our region. Finally, we will present some ideas and research topics we are currently working on at Data Aurora, specifically beneficial for social good purposes.
Global Impact Partnerships

Facebook is taking a partnerships approach to the Sustainable Development Goals, focusing specifically on SDG17: partnership for the goals. This includes unblocking common barriers across all SDGs, like data and technology, but also expanding our global impact partnerships to meaningful drive impact on the SDGs.

Displacement Tracking Matrix: Towards a data-driven response to humanitarian crises

The IOM’s Displacement Tracking Matrix (DTM) is a system to track and monitor displacement and population mobility, provide critical information to decision-makers and responders during crises, and contribute to better understandings of population flows. Here, I shall show some of the data-driven initiatives developed inside DTM, that aims to better support humanitarian response.

Shedding Light on Poverty in the Philippines using Transfer Learning and Open Data

Mapping the distribution of poverty is essential for policy makers and humanitarian organizations for assistance targeted towards vulnerable groups. However, one major challenge is the lack of reliable socioeconomic data, which is highly expensive, time-consuming, and labor-intensive to collect. We tackle this problem by using transfer learning on a variety of data sources including: daytime satellite imagery, nightlights data, human settlement layer, and crowd-sourced geospatial information from OSM.