Weekly News Football

Published on March 25th, 2016 | by Joshua New

0

10 Bits: the Data News Hotlist

This week’s list of data news highlights covers March 19-25, 2016 and includes articles about a government initiative to map public transit in the United States and a partnership between the United States and Europe to share satellite data.

1. Artificial Intelligence Improves Refugees’ Mental Health

A startup called X2AI has created an artificial intelligence program named Karim that can have conversations with users in Arabic via text message to analyze their emotional state and provide recommendations to help improve their mental health. X2AI has partnered with the disaster relief non-governmental organization Field Organization Team to raise awareness about X2AI in refugee camps, where many have experienced emotional trauma. Karim is not intended to replace traditional counselling, but rather serve as a supportive “friend” for refugees in camps where mental-health services are rare.  

2. Mapping Transit Deserts in the United States

The U.S. Department of Transportation (DOT) plans to launch a new initiative this summer to plot the network of bus routes, train lines, and commuter roads around American cities to help urban planners and policymakers improve infrastructure planning and ensure underserved communities can access public transportation. The initiative, called the National Transit Map, will also aggregate data from transit agencies to make it easier for developers to work with this data to create new tools and other maps.

3. Machine Learning Learns to Read Lips

Researchers at the University of East Anglia in the United Kingdom have developed a machine learning model capable of interpreting mouthed words more accurately than human lip readers. The researchers trained their model on images of the shape of the mouth as it makes different sounds, enabling it to read lips without audio cues or contextual information that humans rely on. Lip-reading technology can be useful for people with hearing disabilities and for interpreting audio from video that does not have sound, such as security footage.

4. Sharing Genetic Data from an iPhone

Consumer genetic testing company 23andMe is developing a program to allow customers to pair their genetic data with the health data they collect with their iPhone and share it with medical researchers. By combining genetic information with the health data iPhone apps can log, such as physical activity and medication adherence, researchers can better understand the relationship between lifestyle or environmental factors and medical conditions. Cardiovascular disease and asthma will be the subjects of the first two studies using genetic data.  

5. Using Data Mining to Understand Better City Living

Researchers at the University of Trento in Italy have developed a data mining technique to analyze city environments based on how well a city meets certain conditions defined by urban sociological research that suggests diverse physical environments are conducive to vibrant city life. The researchers combined open data, such as census and land use information, social media data, mobile phone usage, and other data sources in several major Italian cities to analyze how well they fulfilled the conditions for physical diversity. This analysis revealed that certain factors such as high street density and a high number of establishments that are not homes or workplaces are directly correlated with a high level of vibrancy and vitality.

6. Partnering to Observe the Entire Planet

The European Space Agency (ESA) and U.S. Geological Survey (USGS) have partnered to share Earth-observation data from the Sentinel and Landsat satellite networks and make the combined data freely available. Sentinel and Landsat satellites take high-resolution images of the Earth that provide scientists with data to monitor and model land use and changes in the environment, such as polar ice shelf disintegration caused by climate change. By combining data from both satellite networks, researchers will be able to analyze more granular data about the planet.

7, Investigating Concussions in the National Football League

The New York Times has conducted its own analysis of a National Football League’s (NFL) database of player injuries the NFL used to produce research on the prevalence of concussions, and it has found that the NFL omitted at least 100 concussions from its reports on the issue. The data spans from 1996 to 2001 and includes information about a player’s position, symptoms, and recovery time after being diagnosed with a concussion. By comparing the data to other information, such as news articles and team schedules, the New York Times found glaring omissions in the NFL’s research, such as how no Dallas Cowboys players appear in the database, despite quarterback Troy Aikman suffering four concussions during this time period.

8. Predicting When Crowds Get Dangerous

Baidu has developed a machine learning algorithm that can analyze the map searches of its users to predict when a potentially dangerous number of people will congregate in a particular area. Using anonymized historical search data, Baidu’s algorithm can predict the formation of dangerously large crowds up to three hours in advance, which could help authorities intervene before people get hurt. For example, on New Year’s Eve in 2014 in Shanghai, an overcrowded gathering caused a stampede that killed 36 people.

9. Building a Smarter Debt Collection System

Berlin-based startup PAIR has developed an automated system to improve how both creditors and debtors handle paying off debts that relies on learning algorithms to continuously refine debt collection strategies. PAIR’s system, called Realtime Online Settlement Engine (ROSE) analyzes customer payment activity to understand why a person failed to make a payment and develop strategies to help them plan a new payment. For example, if ROSE detects a person recently bought plane tickets before missing a payment, it can guess that the person is out of the country and thus unaware he or she was sent a bill, and determine the best way to contact the person with information about payment options.

10. Delivering Pizzas with Self-Driving Robots

Pizza restaurant chain Domino’s and Australian startup Marathon Robotics have developed a self-driving delivery cart named DRU (Domino’s Robotic Unit) to deliver pizzas to customers. DRU relies on LIDAR sensors navigation technologies similar to those used by self-driving cars to identify and navigate the most efficient route of sidewalks and bike paths to reach a customer. Domino’s only has one DRU so far but is working to expand its use as its ability to operate on sidewalks allows it to bypass heavy traffic.

Image: skeeze.

Tags: , , , , ,


About the Author

Joshua New is a policy analyst at the Center for Data Innovation. He has a background in government affairs, policy, and communication. Prior to joining the Center for Data Innovation, Joshua graduated from American University with degrees in C.L.E.G. (Communication, Legal Institutions, Economics, and Government) and Public Communication. His research focuses on methods of promoting innovative and emerging technologies as a means of improving the economy and quality of life. Follow Joshua on Twitter @Josh_A_New.



Back to Top ↑

Show Buttons
Hide Buttons