Professional Documents
Culture Documents
Altitude (km)
The improved perception provided by the 3
sensing and processing systems, coupled with 2
the improved persistence provided by atmo-
spheric energy harvesting, should enable 1
long-endurance missions that are far beyond 0
the capabilities of current robotic aircraft.
60
Eventually a soaring-capable, autonomous,
CREDIT: (PLOT) A. CHAKRABARTY/DEPARTMENT OF AEROSPACE ENGINEERING, PENN STATE UNIVERSITY. (DATA) G. S. YOUNG, B. J. GAUDET, N. L. SEAMAN, D. R. STAUFFER/DEPARTMENT OF METEOROLOGY/PENN STATE UNIVERSITY
50
mini-UAV that is equipped with a sophisti-
cated sensing system will be able to follow a 40
migrating bird and provide close-up in-flight 30
North (km)
video, as well as in situ atmospheric mea- 20
60
50
surements. Successful completion of such a 10
30
40
COMPUTER SCIENCE
Real-time data on the whereabouts and
S
omething important is changing in human health, guide traffic, and advance the where each picture was taken, and an interface
how we as a society use computers to scientific understanding of human behavior. that observes every incoming and outgoing
mine data. In the past decade, machine- Meanwhile, new algorithms aim to address e-mail and text message. The potential bene-
learning algorithms have helped to analyze privacy concerns arising from data sharing fits of mining such data are various; examples
historical data, often revealing trends and pat- and aggregation (9, 10). include reducing traffic congestion and pollu-
terns too subtle for humans to detect. Exam- To appreciate both the power and the pri- tion, limiting the spread of disease, and better
ples include mining credit card data to dis- vacy implications of real-time data mining, using public resources such as parks, buses,
cover activity patterns that suggest fraud, and consider the data available just to your phone and ambulance services. But risks to privacy
mining scientific data to discover new empiri- company, based on your phone records and from aggregating these data are on a scale that
cal laws (1, 2). Researchers are beginning to those of millions of other individuals who are humans have never before faced.
apply these algorithms to real-time data that going about their daily lives carrying a smart One line of research is based on watch-
record personal activities, conversations, and phone—a device that contains a Global Posi- ing where people are, where they are head-
movements (3–8) in an attempt to improve tioning System (GPS) sensor locating you to ing, and when. Anonymous real-time loca-
within a few meters, an accelerometer that tion data from smart phones are already being
Machine Learning Department, Carnegie Mellon University,
detects when you are walking versus stationary, used to provide up-to-the-minute reports
Pittsburgh, PA 15213, USA. E-mail: tom.mitchell@cs.cmu. a microphone that detects both conversations of traffic congestion in many urban regions
edu and background noises, a camera that records through services such as Google Maps (11).