DataKind/WorldBank Big Data Exploration against poverty and fraud
About 150 data scientists, civic hackers, visual analytics savants, poverty specialists, and fraud/anti-corruption experts made DataKind/WorldBank DataDive at Washington DC a success. See some highlights.
DataKind and World Bank organized a DataDive in Washington, DC on Mar 15-17, 2013, and it attracted about 150 50 data scientists/hackers, experts in visual analytics, and specialists in fighting poverty, fraud, and corruption.
Here are some of the projects that volunteers worked on
- Predicting Small-Scale Poverty Measures from Night Illumination - can freely available satellite imagery, showing average nighttime illumination, serve as a reasonable poverty measurement proxy? (yes, found that light information was predictive of poverty in 2001 with 2001 light data and 2001 census information.)
- Analyzing the World Bank's Project Data for 'Signals' - do successful or unsuccessful projects (or projects reporting corruption and the ones that don't) share any characteristics?
- Social networking analysis for risk measurement - can you forecast project risk using social networking analysis tools?
- World Bank Supplier Profiles - can the Bank and other agencies include publicly available data to gain a broader, more comprehensive understanding of their suppliers and use the information as proxies for risk management?
- Measuring Socioeconomic Indicators in Arabic Tweets - can Twitter data help you understand socio-economic trends in countries?
- Can you use simple heuristic auditing to sniff out discrepancies in expenditure data - what do you do when you have the information but don't know if it contains signals about potential fraud and corruption related risk?