10 Bits: The Data News Hotlist
This week’s list of data news highlights covers June 6-12, 2015 and includes articles about the National Day of Civic Hacking and how a new how San Jose is connecting itself to the Internet of Things and how artificial intelligence is helping developers make better apps.
The city of Angers, France opened a new government-funded Internet of Things incubator called Cité de l’Objet Connecté (“City of the Connected Object”) in an effort to cultivate more high-tech manufacturing in the country. The incubator will assist startups and established companies alike with the development of new connected technologies from conception to production, and includes on-site fabrication facilities. Government subsidies and loans currently make up half of the incubator’s budget, but if successful, the program should be self-sustaining within three years.
106 cities and communities held events on June 6 for the National Day of Civic Hacking, an annual hackathon designed to encourage government and public collaboration, organized by Code for America, a nonprofit organization devoted to improving government use of technology. Many of the participants developed apps and software tools designed to improve municipal open data initiatives in accordance with the event’s theme, “Principles for 21st Century Government.” For example, in California, a team built an app using data from California’s Health and Human Services Agency’s open data portal that aggregates and maps information about nearby resources for low- and medium-income pregnant women, new mothers, and children supported by Sacramento County’s Women, Infants, and Children health and nutrition program.
Researchers from Stanford University analyzed 16 million electronic health records on 2.9 million people from multiple databases to reveal a potential connection between a popular type of heartburn medication and an increased risk of heart attack. While more research is needed to confirm the linkage, the potential increased risk is substantial—from 16 to 21 percent. Importantly, the researchers included doctor’s notes in their analysis of medical records—unstructured data that has been historically difficult to process.
South Korean officials are monitoring mobile phone location data of citizens under quarantine after an outbreak of the Middle East Respiratory Syndrome (MERS) virus in late May. The data will help officials prevent the spread of the disease, which has a death rate of 38 percent, and avoid situations similar to when a South Korean man broke quarantine in May to travel to China, where he then tested positive for MERS. The quarantine and monitoring apply to approximately 2,000 people in the country, and health officials expect this isolation will halt the outbreak, which has killed six and infected 87 thus far.
Researchers at the National Institutes of Health and the Department of Veterans Affairs (VA) have developed a new database of VA patient records and an algorithm to comb these records, resulting in the identification of groups of patients with suicide rates 80 times higher than the VA average. Many of these high-risk cases have previously gone unidentified by the current system, which relies on medical staff to flag patients. The VA is still determining how this new information will impact treatment, but this substantially improved method of risk assessment should prove helpful for targeting care to patients that need it most.
A new website developed by Eastern Washington University will aggregate federal and state open datasets to guide informed decisionmaking in Benton and Franklin Counties in Washington. The website, called Benton-Franklin Trends, pulls together data that is already publicly available but disparate and sometimes hard to access into a single portal and groups datasets into categories such as public safety, education, and transportation. The website’s supporters, which include the university and local community groups, also expect the economic and workforce data will entice businesses to come to the region, as this data has not previously been readily available.
India has begun work on a new system to collect comprehensive data on national employment to better understand the country’s economic health. Though the Indian government has several programs that collect employment data, its surveys are either conducted too infrequently or examine only a portion of the economy. Existing methods also fail to accurately account for the large amounts of informal employment in India. Once complete, the new program will conduct national comprehensive surveys on an annual basis and examine urban areas every quarter to provide more accurate economic data for policymakers.
Facebook has made Infer, its artificial intelligence software debugging tool, freely available to help developers improve code in their apps. Infer relies on an advanced version of an artificial intelligence technique called static program analysis that can rapidly process thousands of lines of a program’s code to catch and fix potential errors with a success rate of 80 percent. The developer of Infer hopes that by making the tool open source, industry and academic researchers can collaborate to improve the effectiveness of this type of analysis.
The Transport System Catapult, a UK government-backed organization designed to support innovation in transportation technology, has published a repository of datasets to support groups developing advanced transportation technologies like driverless cars. The repository pulls from over 200 sources to compile data on topics ranging from the environment to freight transportation. The catapult hopes that consolidating this data, most of which is free to access, will spur innovation by providing transportation technology researchers with easy access to useful information they might not otherwise have exposure to.
The Deep Space Climate Observatory satellite, developed by NASA, the National Oceanic and Atmospheric Administration, and the U.S. Air Force, has begun its orbit around the sun to collect data that could give advanced warning of potentially harmful solar storms. Solar storms could cost up to $2 trillion in damages to power grids, telecommunications equipment, and other infrastructure. The advanced warning could help scientists warn at-risk populations ahead of time to minimize the damage.
Image: flickr user Ryan.