The Bill and Melinda Gates Foundation has published a new report on Using mobile big data for development. Too often basic information on the poor – who is poor, where do they live and move, how do they manage their social and financial lives? – is scarce, while the costs associated with being underknown are significant. Without formal financial histories, creditworthy individuals cannot access loans when needed most, vaccine workers cannot determine what percent of a region they have immunized from a disease, and relief organizations cannot anticipate where people move when catastrophic events occur. Overall, the quality and efficiency of providing services to the poor suffers.
Much has been written about Kenya’s m-money system. Here the Economist highlights a Gates Foundation paper that highlights an aspect that has not been much written about, the need to balance e money and real money in the hands of the retailers. There are many elements to a successful mobile-money scheme: the right technology, simple marketing, partnerships with banks, support from regulators. But keeping it all going are people like Gaudencia, moving bundles of cash around, on buses and in vans, behind the scenes.
Assume a scenario where among the chief complaint strings of two unrelated patients in the same District on the same date there was a mention of bloody stools in pediatric cases. The multiple mentions of “bloody stools” or “pediatric” might not be surprising, but the tying together of these two factors, given matching geographic locations and timings of reporting, is sufficiently rare that seeing only two such cases is of interest. This was precisely the evidence that was the first noticeable signal of the tragic Walkerton, Canada, waterborne bacterial gastroenteritis outbreak caused by contamination of tap water in May 2000. That weak signal was spotted by an astute physician, not by a surveillance system. Reliable automated detection of such signals in multivariate data requires new analytic approaches.