Paper Reading #17- Privacy risks emerging from the adoption of innocuous wearable sensors in the mobile environment

Title: Privacy risks emerging from the adoption of innocuous wearable sensors in the mobile environment
Reference Information:
Andrew Raji, Animikh Ghosh, Santosh Kumar, and Mani Srivastava. "Privacy risks emerging from the adoption of innocuous wearable sensors in the mobile environment". CHI '11 Proceedings of the 2011 annual conference on Human factors in computing systems ACM New York, NY, USA ©2011. ISBN: 978-1-4503-0228-9.
Author Bios:
Andrew Raji- Works at the University of South Florida as a tenure-track Assistant Professor. He worked as a postdoctoral fellow from 2009-2010 and coordinated both the AutoSense and FieldStream projects.
Animikh Ghosh-  Joined SETLabs, Infosys after completing his M.S. in Comp Sc. from University of Memphis, Tennessee, USA. Now a research assistant under Dr. Santosh Kumar.
Santosh Kumar- I lead the Wireless Sensors and Mobile Ad Hoc Networks (WiSe MANet) Lab at the University of Memphis.
Mani Srivastava- Is a professor at UCLA in the engineering department. He received both the M.S. and Ph.D. degrees from the University of California, Berkeley, in 1987 and 1992, respectively.
Summary:

• Hypothesis: If commercially-available mobile devices are going to continue to be worn, used, and distribtued, then their associated privacy leak concerns should be addressed immediately in order to prevent unintended information leaks about users through inferences of collected information from such technologies.
• Methods: The authors studied groups of people to get feedback about privacy concerns.These groups of people used a technology called AutoSense. One group of people had no personal ties to data presented later on in the study while the other group was tested twice. They were tested after 3 days of using the device, and then again after all of the data was collected and presented back to them including inferences about each of them using the data from the technology to see if their concern level changed. The authors used 6 main points to focus their conclusions: measurements, behaviors, contexts, restrictions, abstractions, and privacy. Measurements are the raw data sets captured by a sensor. Behaviors are the inferred actions taken by the data producer with associated measurements. Context is any information that can be used to imply the situation of the behavior. Privacy threats are those that come with the data producer's identity being tied to the information. Restrictions remove data from a dataset before it can be shared to recude privacy threats. Abstractions provide a middle ground between complete restriction and complete sharing of data in order to render the device useful.
• Results: The study found that users are most concerned with the release of information concerning conversation and commucing patterns. Obviously, the group with no personal stake in the data couldn't have a strong concern to the data from the second group of users using the AutoSense because they said that it didn't really bother them because it wasn't their own personal information. Upon data presentation sessnions, the authors also found (not surprisingly) that participants' concern levels rose as more and more information was released at a time. For example, if just timestamps of events or just events rather than a timestamp with an associated event all released together. This way, you couldn't really piece together information as easily as you could if the event and time were right there next to each other. The authors found that abstractions play a large role in determining the level of privacy threats.
• Content: The authors conducted these studies to raise general awareness about privacy issues associated with general or specific-purpose mobile devices most people carry around today. They found that it is one issue to share personal information with someone you are directly talking to, but it is another issue entirely to share information across a wide network ("the web") so the general public may go access it.

Discussion:
This was an interesting read. It made me think about what all information I have released to the general public in the past and I will definitely be a little cautious when making decisions in the future. The authors had access toa LOT of information about the participants just because they were wearing a sensor around for 3 days. The inferences that can be made from the measurement collections are very strong as I have learned. These issues should be addressed more widespread when companies create such technologies, in my opinion. Based upon the results of their study, I believe the authors definitely achieved what they had set out to do initially. They are just trying to raise awareness about the things that the "leak" unintentionally to the "public", much like when you are visiting sites online and you "save your password" for a site.