The Global Pulse Project Series showcases data innovation projects carried out with our partners across the network of Pulse Labs in New York, Jakarta and Kampala. The projects have been documented as case studies to enable others to adapt and build upon the approaches. [Download press release – PDF]
Projects were in collaboration with UN agencies, governments, academia and private sector partners and demonstrate how new sources of digital data could support development planning, and inform targeting of humanitarian response efforts.
With all eyes on the new Post-2015 framework, and the call for a "data revolution," this collection of case studies attempts to show a path from theory to action – using data science and innovative partnerships to approach development and humanitarian challenges differently, and offering new ways to not only measure but achieve sustainable development goals.
Summaries & Downloadable Case Studies:
Mapping the Risk-Utility Landscape of Mobile Data for Development & Humanitarian Action [PDF]
This project assessed the impact that aggregating mobile data to protect privacy has upon the utility of the data for transportation planning and pandemic control and prevention. The proposed methodology allows for determining what level of data aggregation is the minimum required to adequately protect individual privacy while preserving its value for policy planning and crisis response. This project was done in collaboration with the MIT Connection Science. (Project webpage)
Data Visualization & Interactive Mapping to Support Response to Disease Outbreak [PDF]
This project developed interactive data visualization tools that were used during a typhoid outbreak in Uganda to analyze dynamic information about case data and risk factors in support of the national task force managing the outbreak. This project was done in collaboration with WHO and the Ugandan Ministry of Health. (Project webpage)
Understanding Public Perceptions of Immunisation Using Social Media [PDF]
This project extracted and analyzed tweets related to vaccines and immunization in Indonesia. Findings included the identification of perception trends including concerns around religious issues, disease outbreaks, side effects and the launch of a new vaccine. This project was done in collaboration with the Ministry of Development Planning and the Ministry of Health in Indonesia, UNICEF, and WHO. (Project webpage)
Understanding Immunisation Awareness and Sentiment through Analysis of Social Media and News Content [PDF]
This multi-country study analyzed perceptions about immunization from multiple social media channels and news sources in India, Kenya, Nigeria and Pakistan. The project shows how methods including sentiment analysis, topic classification and network analysis can be used to support public health workers and communication campaigns. (Project webpage)
Analyzing Attitudes Towards Contraception & Teenage Pregnancy Using Social Data [PDF]
This study analyzed public Facebook data to gain real-time awareness of changing trends in attitudes about contraception methods and family planning discourse among Ugandans. This study was done in collaboration with UNFPA. (Project webpage)
Analysing Social Media Conversations to Understand Public Perceptions of Sanitation [PDF]
This study produced a baseline of Twitter activity on topics related to sanitation. Findings revealed a large proportion of conversations related to cholera and increasing public engagement around gender issues and its intersection with sanitation. This study was done in collaboration with the UN Millennium Campaign and WSSCC. (Project webpage)
Supporting the Post-2015 Development Agenda Consultations Using U-Report [PDF]
This project supported Uganda’s Post-2015 consultation process on national priorities by analyzing 3.1 million messages from UNICEF’s citizen reporting platform. A data visualization tool was developed to map and classify the views of Ugandan youth around development topics. This project was done in collaboration with UNDP and UNICEF. (Project webpage)
Using Twitter to Understand the Post-2015 Global Conversation [PDF]
This project developed a real-time on-line dashboard that shows the volume and priority topics discussed in social media around the world related to the Post-2015 development agenda. This project was done in collaboration with the UN Millennium Campaign and DataSift. (Project webpage)
Feasibility Study: Supporting Forest and Peat Fire Management Using Social Media [PDF]
This feasibility study explored the relationship between Twitter trends during major forest fires or haze events, in relation to on-the-ground events in Indonesia. The study aims to show how social media signals could support emergency response management. This study was done in collaboration with UN Office for REDD+ Coordination in Indonesia. (Project webpage)
Using Twitter to Measure Global Engagement on Climate Change [PDF]
This project developed a real-time social media monitor to explore online discourse about climate change in support of the United Nations Climate Summit in 2014. The publicly accessible monitor analyzed tweets in English, Spanish and French on a daily basis to show the volume and content of tweets about climate change across a range of topic areas such as economy or energy. (Project webpage)
Feasibility Study: Identifying Trends in Discrimination Against Women in the Workplace in Social Media [PDF]
This feasibility study tested a method for filtering keywords from public tweets related to discrimination against women in the workplace, identifying some topics with significant volume of discussions – such as discriminatory job requirements. This study was done in collaboration with ILO and Crimson Hexagon. (Project webpage)
Estimating Migration Flows Using Online Search Data [PDF]
This study explored whether online search data could be analyzed to understand migration flows and produce a proxy for migration statistics, using Australia as case study. This project was done in collaboration with UNFPA. (Project webpage)
Mining Citizen Feedback Data for Enhanced Local Government Decision-Making [PDF]
This project deployed data analysis and visualization tools to structure and combine data from the Indonesian national citizen-reporting complaint system and a local SMS-based feedback system (representing active citizen complaints), together with public Twitter posts (representing passive opinions). This project was done in collaboration with the NTB Provincial Government in Indonesia. (Project webpage)
Using Twitter Data to Analyse Sentiment on Fuel Subsidy Policy Reform in El Salvador [PDF]
This study analyzed the temporal evolution of the content and sentiment of tweets related to a fuel subsidy reform in El Salvador. The study demonstrated that public opinion as expressed in social media could complement existing methodologies for opinion mining based on survey data. This study was done in collaboration with the World Bank. (Project webpage)
Feasibility Study: Analysing Large-Scale News Media Content for Early Warning of Conflict [PDF]
This feasibility study explored how data mining of large-scale online news data could complement existing tools and information for conflict analysis and early warning. Taking Tunisia as a test case, analysing news media archives from the period immediately prior to and following the January 2011 government transition, the study showed how tracking changes in tone and sentiment of news articles over time could offer insights about emerging conflicts. This project was done in collaboration with UNDP. (Project webpage)
Using Mobile Phone Activity For Disaster Management During Floods [PDF]
This project combined the analysis of mobile phone activity data with remote sensing data during severe flooding in the Mexican state of Tabasco as a method to inform emergency management response. This project was done in collaboration with WFP, the Government of Mexico, the Universidad Politecnica de Madrid and Telefonica Research. (Project webpage)
Feasibility Study: Crowdsourcing High-Frequency Food Price Data in Rural Indonesia [PDF]
This feasibility study used crowdsourcing to track commodity prices in near real-time in areas where the availability of other data sources is limited. The study involved recruiting a trusted network of local citizen reporters to submit food price reports via mobile phones. This project was done in collaboration with FAO, WFP and Premise. (Project webpage)
Analysing Seasonal Mobility Patterns Using Mobile Phone Data [PDF]
This project quantified seasonal mobility of populations in different regions of Senegal, based on analysis of anonymised mobile phone activity data. This project was part of the Orange D4D Challenge, and done in collaboration with WFP and the Universidad Politecnica de Madrid. (Project webpage)
Nowcasting Food Prices in Indonesia Using Social Media Signals [PDF]
This project investigates how people's self-reporting of commodity prices through Twitter can be used to provide real-time price indicators. The project showed that the prices reported on Twitter were closely correlated with official food prices. This project was done in collaboration with the Indonesian Ministry of Development Planning, WFP and the Korea Advanced Institute of Science and Technology. (Project webpage)
Using Mobile Phone Data and Airtime Credit Purchases to Estimate Food Security [PDF]
This study assessed the potential use of mobile phone data as a proxy for food security. Results showed high correlations between airtime credit purchases and survey results referring to consumption of several food items. In addition, models based on anonymised mobile phone calling patterns and airtime credit purchases were shown to accurately estimate multidimensional poverty indicators. This project was done in collaboration with WFP, Université Catholique de Louvain and Real Impact Analytics. (Project webpage)
