Wednesday, November 4, 2009

Social Tagging to Verify Identity

Because the anonymity of the web might allow a major chunk of social network users to be dishonest, an algorithm being developed by a Duke graduate student is trying to estimate a given user's credibility by asking their friends if they're legit.

Michael Sirivianos, a PhD student in Computer Science presented FaceTrust in a talk on Tuesday.

It's a unique system he and his fellow researchers developed that uses a person's Online Social Network (OSN) and reputation among friends to verify how trustworthy they may be. He calls it "relaxed and attribute-based credentials."

In the talk entitled "On the Internet, 'Am I really not a dog?' " Michael explained the complete concept behind their system and how it could be an essential step in assessing the personality of users on social networks.

"An online world without identity credentials makes determining who to believe very difficult," Michael said.

Sites like Amazon, eBay, dating sites, Craigslist etc., might "simply ask you a question, like, if you are over 18 or not, and if you say yes, there is no other checking mechanism to determine if you are speaking the truth or not," Michael said.

The approach that Michael and his team took in solving this problem is inspired by the wisdom of crowds-- the power of user feedback. FaceTrust employs "crowd vetting," i.e., using feedback from friends to determine whether an online user's statement or assertion may be credible.

"Online Social Network users tag/vote for their friends' verifiable identity assertions. These OSN providers issue credentials on the user's assertions," Michael explained.

"We do tagging using a concept called social tagging. Currently, we use Games With a Purpose to assess the credentials. Basically, a combination of fun and serious (useful) assertions are presented to the user's friend, and they vote for a yes or no," Michael added.

Since FaceTrust utilizes OSN and friend feedback as reliable sources, it is successful in providing probabilistic assurances of a user's credibility.

"Of course a number of concerns come up- there can be dishonest users, credentials can be forged, and maybe the users don't even tag at all. Also, we need to preserve the privacy of taggers and preserve anonymity of users that present credentials."

To ensure that the feedback from only honest users is weighted in the algorithm, only friends of that user on that network can tag a user. Friending can be seen as a form of trust with high probability.

The algorithm also assumes that most honest users have friends who will not tag their honest assertions about themselves as false.

"To further ensure the fact that the feedback we collect over a large range is reliable, we employ trust transitivity via a method called Trust Inference. We use history to determine similarity. It is observed that honest friends have a history of tagging attributes about a person similarly," Michael explained.

The complete paper, entitled "FaceTrust: Assessing the Credibility of Online Personas via Social Networks" can be found here.

Michael Sirivianos completed his B.S. in Electrical and Computer Engineering at the National Technical University of Athens, and M.S. in Computer Science and Engineering at UCSD. His other projects include Free-riding in BitTorrent networks, Loud and Clear(L&C), and Dandelion.

Sunday, November 1, 2009

Who Decides Who Lives and Dies?

As an audience filed into a Duke Hospital auditorium last week to hear Philip M. Rosoff talk about the ethical difficulties of apportioning scarce resources to fight a pandemic, a line of people waiting for flu shots in the hallway outside stretched out of sight.

Rosoff, a professor of Pediatrics and Medicine at the Trent Center for Bioethics, Humanities and History of Medicine, raised the same question as the flu shot queue: “How does one equitably distribute limited resources to as many as could benefit from them?”

Who will live and who will die? Who decides? As a result of improved technology and effectiveness of medicine, healthcare workers must make increasingly difficult decisions.

“Most of the dilemmas facing medicine in 2008 did not exist in 1918,” Rosoff said. “You either got better, or you died.”

“Our expectations of what modern scientific medicine can provide are totally different today. We expect ‘miracles’ every day and are vastly disappointed and even shocked when medicine doesn’t deliver,” Rosoff said.

With scarce resources in the emergency room, how would you choose between a 21-year-old honor student and an 86-year-old nursing home resident? Not a difficult choice for most in the audience. But how would you choose between a 21-year-old honor student and a 40-year-old mother of three young children? Between two 21-year-old honor students? Or between an honor student and the child of a colleague? The audience grew solemn as he went down the list.

“Everyone was happy to knock off granny,” Rosoff joked.

Planning for a potential epidemic reveals some critical problems. “Even during a regular flu season, about 100,000 medical ventilators are in use. In a worst-case human pandemic, the country would need as many as 742,500,” Rosoff said.

Duke Hospital has a supply of 150 ventilators, of which 90% are in use or in repair in any given time. Of those, a number are suitable only for infants or young children. The Strategic National Stockpile has 4,900 ventilators in strategic locations across the country and hospitals nationwide have about 105,000 ventilators.

“People will still get cancer, will still have heart attacks, acute appendicitis and premature babies and thus need hospitalization: what happens to them if we have a pandemic? Do we send the 65-year-old cancer patient home so we have a bed for the 35-year-old with influenza?”

Rosoff said some allocation systems have been proposed:
  • Minor adaptations of standard emergency department triage systems for assigning ventilators
  • The “life cycle principle” for rationing vaccine, privileging the young over the old (with the idea that each person should have an opportunity to live through the different cycles of life)
  • A variation of the organ transplant distribution system, which is already widely accepted
“We should not ration care based on morally arbitrary and irrelevant features about people, such as skin color, ethnicity, religion, age, gender, etc. ... Social worth -- however you feel about people -- should not play into it.”

“Decision making and planning should be transparent and reasons for decisions should be available and defensible,” Rosoff said. “We should apportion care based upon substantiated and justified evidence of its efficacy. We can make societal decisions that anyone over a certain age, or anyone with certain types of disabilities or people not in the country legally will not be eligible for curative care, but this has to be a consensus decision.”

Most importantly, “we must be prepared and have thought through many of these issues ahead of time.”

His proposal for North Carolina, once a healthcare emergency has been declared and resources are scarce:
  • Initiate the Hospital Emergency Incident Command System to coordinate resource allocation.
  • Pre-existing conditions will exclude some patients from consideration for life-preserving care. For example: severe cognitive impairment, heart failure, liver disease, metastatic cancer with less than a year life expectancy and severe burns.
  • Life preserving resources will be allocated to patients based upon a modified SOFA score with age component (younger patients privileged).
  • There will be an appeals process available to patent families 24/7 to contest a decision.
  • Healthcare workers with influenza would not be privileged to receive resources ahead of other patients (colleagues and coworkers do not jump to the head of line -- this is to bestow legitimacy for the public)
“We must reaffirm our moral commitments to serve as a guide for our actions,” Rosoff said. “Hard choices must be public, acceptable and justifiable.”