Weekly News The London Underground

Published on August 15th, 2014 | by Travis Korte

0

10 Bits: The Data News Hot List

This week’s list of data news highlights covers August 9-15 and includes articles about an online tool that predicted the ongoing Ebola outbreak before the World Health Organization and a partnership between the Michael J. Fox Foundation and Intel to research using wearable technology to improve monitoring for Parkinson’s disease.

1. Canada Joins Rare Disease Data Sharing Consortium

The Canadian government has become the first country in the western hemisphere to join the international Internet-based rare disease data and information portal Orphanet. The portal, which aims to help patients with rare diseases learn about their conditions and get access to medical help, will also help expedite data sharing among clinicians. Orphanet has already been adopted in numerous European countries.

2. Reducing Credit Card Fraud at the Gas Pump

Chevron is using Visa’s predictive analytics software to reduce credit card fraud at gas stations. The software, Visa Transaction Advisor, can identify stolen and counterfeit cards by analyzing about 500 variables, including past transactions and location. Chevron reports that a pilot program to use the software cut gas pump fraud by 23 percent over a 2-month period and that it plans to expand the use of the software at its gas stations nationwide.

3. Online Tool Predicted Ebola Outbreak Before WHO

An online disease surveillance tool called HealthMap spotted the ongoing West African Ebola outbreak nine days before experts at the World Health Organization announced it. HealthMap gathers data from social media, local news, government databases, and other sources to detect and map disease activity. The tool also lets users graph the growth of diseases over time and search for diseases by source, location, and even vector species.

4. Genetic Analysis Helps Predict PTSD

Researchers from the Mount Sinai School of Medicine published a paper this week demonstrating how genetic analysis could be used to predict which individuals are likely to develop post-traumatic stress disorder (PTSD). To reach their conclusions, the researchers exposed rats to the smell of cat urine and measured expressions of 22,500 genes in the rats that exhibited symptoms of trauma. They found that the traumatized rats expressed certain genes differently and thereby had slightly different brain chemistry from the unaffected rats. The researchers hope their work will help future researchers develop treatments for PTSD that target gene expression.

5. Siri Creators Team Up For New Smart Assistant

Members of the team that worked on Apple’s Siri digital assistant have founded a new startup called Viv, which aims to create an even smarter intelligent assistant system that will allow users to create much more complex queries. The company offers an example query—”On the way to my brother’s house, I need to pick up some cheap wine that goes well with lasagna”—and shows the process by which Viv might recommend wine stores along the route to the brother’s house that carry certain kinds of wine.

6. Transport for London, Hotbed of Data Innovation

Transport for London (TFL), the city’s transportation authority is using data in a variety of ways to move people through the city more safely, quickly, and cheaply. For example, the agency has a network of cameras and roadside sensors that allow cameras to change traffic light patterns and optimize traffic, and it will soon pilot a similar approach for optimizing pedestrian flow. In addition, TFL has developed forward-looking reports anticipating infrastructure needs for a future full of driverless buses. It releases much of its data openly, and users have developed an array of apps to help passengers navigate through the city using various forms of transit.

7. Michael J. Fox Foundation, Intel Partner on Parkinson’s Wearables

This week, the Michael J. Fox Foundation and Intel announced a research partnership to improve Parkinson’s disease monitoring using wearable computing and data analytics. The partnership is already in progress, with Intel data scientists analyzing information on medication intake, symptoms, and other factors from test patients. In the next phase of the project, the organizations will develop a mobile app for patients to record their own activity and send it to doctors.

8. Machine Learning Algorithm Predicts Heart Attacks Better than Doctors

Researchers at Carnegie Mellon University have developed a machine learning algorithm to predict heart attacks in high risk patients. The algorithm, which looked at 72 variables from around 133,000 patients, was more than twice as accurate at predicting cardiac arrest events as traditional methods, and in some cases was able to predict heart attacks up to four hours before an event. Next, the team hopes to make the system more attractive to hospitals by refining its models using even more data, so that predictions can be made with greater confidence in the future.

9. Data-Driven Microtargeting in the Renewable Energy Sector

Renewable energy company Ethical Electric is using data analysis methods developed for political campaigns to pinpoint potential customers. The company, which operates in seven states and Washington, D.C., merges data from a variety of public data sources to identify, down to the household level, where it might find people interested in buying renewable power. As a result, the company has been able to reduce the number of prospective customers it contacts more than tenfold.

10. Automatic Photo Editing Algorithm Can Change Seasons, Weather

Brown University researchers have developed an algorithm that can automatically alter images to simulate changes in “transient attributes,” i.e., features like weather, season, time of day that affect the holistic appearance of the image. The researchers also studied how well people respond to images altered with the algorithm and found that 70 percent of those surveyed preferred the algorithmically generated version to a human-made one. The algorithm’s creators hope their work will help make image editing more accessible to non-expert users.

Photo: Creative Commons / Tom Page 

Tags: , , , , , , , ,


About the Author

Travis Korte is a research analyst at the Center for Data Innovation specializing in data science applications and open data. He has a background in journalism, computer science and statistics. Prior to joining the Center for Data Innovation, he launched the Science vertical of The Huffington Post and served as its Associate Editor, covering a wide range of science and technology topics. He has worked on data science projects with HuffPost and other organizations. Before this, he graduated with highest honors from the University of California, Berkeley, having studied critical theory and completed coursework in computer science and economics. His research interests are in computational social science and using data to engage with complex social systems. You can follow him on Twitter @traviskorte.



Back to Top ↑

Show Buttons
Hide Buttons