This week’s list of data news highlights covers December 5 – 11, 2015 and includes articles about how Australian universities are making it harder for graduates to lie about their degrees to employers and how Google search data can help researchers study the spread of sexually transmitted diseases.
The U.S. Department of Transportation (DoT) has announced the Smart City Challenge to spur the development of a mid-sized smart city that demonstrates the ability of connected technologies and data to reduce congestion, improve transportation safety, support the economy, and benefit the environment. Cities can submit bids to participate in the Smart City Challenge until February 2014, and the winning city will receive $40 million to implement its smart city plans. Additionally, philanthropic investment company Vulcan Inc. has agreed to award an additional $10 million to the winning city if its projects include developing infrastructure to support electric vehicles.
A group of Australian universities have launched the Digital Student Data Project (DSDP), an initiative to build a database of academic records to make it easier for employers to verify job applicants’ education credentials. DSDP will have the added benefit of making it easier for students to move between academic institutions without the hassle of transferring paperwork to prove their academic history. DSDP will only grant employer access to academic records if the student or graduate explicitly gives the employer permission. DSDP plans to eventually expand the database to include academic records from other countries, including the United States, China, India, and the UK.
Open Knowledge International, a non-profit organization, has released the 2015 Global Open Data Index assessing the state of open government data in 122 countries. The Global Open Data Index evaluates how well a country provides various categories of important data, such as national statistics and government spending, to the public based on factors such as whether or not the data is freely available, machine readable, timely, or if it exists at all. For the 2015 Index, Taiwan topped the list of countries with an openness score of 78 percent.
Police arrested a woman in Florida for a hit-and-run after data from her car proved she was involved in an accident that hospitalized another woman, contradicting her denial. The woman had struck another car from behind and sensors on the car detected the accident, causing it to automatically call authorities via the woman’s phone as part of a feature called 911 Assist. The woman initially ignored the dispatcher, causing the car to pull over and share its location with authorities on the chance that the driver was incapacitated. The woman then denied to the dispatcher she struck another car, but authorities were able to find her thanks to the car’s location data and identified extensive vehicle damage and paint transfer, proving she had hit another car and tried to flee the scene.
Australian Prime Minister Malcolm Turnbull has released the government’s Public Data Policy Statement, defining data held by the government as a “strategic national resource.” The new policy commits to the practice of open by default for non-sensitive government data, focuses on expanding the use of government data by collaborating with researchers and businesses, and requires that data be made available in machine-readable formats as well as utilize open standards and licenses. The policy is part of the government’s new National Innovation and Science Agenda which aims to advance government research and support startups.
Reality Editor, a new smartphone app developed by the Massachusetts Institute of Technology’s (MIT) Fluid Interfaces Lab, which studies human-computer interaction, allows users to easily link and program the various smart devices in their home. The app allows users to easily link the functionality of different devices. For example, users can link a smart lightbulb and television to automatically dim the lights when the television turns on. Reality Editor is free to download, and developers are free to create software for it to work with smart devices from different manufacturers.
Researchers at University of Illinois at Chicago are mining Google search data to develop analytical models that can forecast outbreaks of sexually transmitted diseases (STD) based on real-time search data for phrases related to various diseases, such as particular symptoms. Researchers can analyze search keyword trends by specific locations and develop a clearer understanding of where disease is spreading before public agencies officially report verified cases. The researchers are focusing on STDs in particular because of their stigmatized and personal nature, which makes individuals less likely to provide data about their disease in other ways useful to epidemiologists, such as tweeting about their congestion when they have a cold.
Researchers at New York University, the University of Toronto, and MIT have developed an artificial intelligence algorithm capable of recognizing handwritten characters as accurately as a human can after analyzing just one example. Until now, even the best machine-learning algorithms needed to analyze thousands of examples to reliably distinguish between different characters. The researchers’ technique differs from other algorithms in that it attempts to mimic the way adults, who already know how to read and write, learn and recreate new characters, whereas other techniques could be described as mimicking the way children learns to recognize a new character as they learn how to read in general.
Startup Descartes Labs has developed a method of creating a live map of Earth unobscured by cloud cover, which can make it easier for scientists to monitor changes on the Earth’s surface, such as changes in vegetation and water features. The company’s software stitches together satellite images of the entire planet from government sources and automatically edits out cloud coverage. With an unobstructed view of the Earth, Descartes Labs claims it can forecast changes more accurately than organizations such as the U.S. Department of Agriculture, which can be valuable for farms and agriculture companies.
The U.S. Centers for Medicare and Medicaid Services (CMS) has published performance scores for individual physicians that have accepted Medicare reimbursements for routine screenings and preventative care for common conditions such as diabetes. CMS only published scores for doctors that have volunteered data about practices such as how reliably they provide follow-up care and how frequently they verify a patient’s medications to avoid potential problems. In 2016, CMS will penalize doctors that fail to report this data.