Evaluation of Human Altruism with DTN based data forwarding: Difference between revisions

Jump to navigation Jump to search
(Created page with "== Details == {{Project Description |supervisor=[http://www.net.informatik.uni-goettingen.de/people/sufian_hameed Sufian Hameed] |duration= 6 months |type= Bachelor Thesis |statu...")
 
No edit summary
Line 8: Line 8:


== Description ==
== Description ==
The explosive growth in the unsolicited email (spam) in the past decade [1] has made it impossible for email communications to function without spam protection/filtering. Currently, spam emails have largely outnumbered legitimate ones, increasing from 65% in 2005 to 89% (262 billion spam messages daily) in 2010. Despite that researchers and practitioners have developed and deployed a broad variety of systems intended to prevent spam; it remains a pressing problem of large scale.
One possible interesting topic as a follow up is about evaluation of human altruism, which is very related to social network you studied in the LENS project. And can also use the Goose as experimental platform. So actually we more of less with all the codes ready and just need to change some of the user interfaces and then we can run the experiment. It is a very interesting topic and can have big impact and most important it will not take much time.
The spam protection systems used today only filter spam from the user’s inbox (i.e. recipient’s edge), but the spam already travels the network, and provokes non-negligible cost to network operators in terms of bandwidth and infrastructure. On the other hand, content-based filtering [3, 4], one of the most widely adopted defense mechanism, has turned spam problem into false positive and negative one. In consequence, this makes email delivery unreliable.


In recent years several techniques [5, 6 and 7] have been presented using social networks to fight spam. Unfortunately their services are only limited within the social network of an email user. At Computer Netwoks groups, in collaboration with Deutsche Telekom labs, we are actively working on LENS, a new spam protection system, which leverages anti-social networking paradigm based on an underlying trust infrastructure to both extend spam protection beyond a user’s social circle and fundamentally prevent the transmission of spam across the network at the first place.  . The key idea of this paradigm in LENS is to select email users called Gatekeepers (GKs), from outside the user’s social circle and within pre-defined social distances. Unless a GK vouches for the emails of potential senders from outside the social circle of a particular recipient, those e-mails are prevented from transmission.
What we are trying to do about testing altruism or trust or incentive etc are very related to what is happening in the field of behaviour economy. People in that research field usually use games (popular prisoners' dilemma, dictator game) to test and observe the behaviours  of the participants when they encountering different choices. But these kind of games are very limited and artificial and I think many of the conclusion are very unreliable, for example from the dictator game, you may observe that a lot of participants are willing to share their money with their partners, but we cannot draw the conclusion that people are altruistic from it. The reason is that usually you recruit students to participate in the game  (usually happened in academic research), and under your supervision/observation (you are the professor), the students will tend to behave nicely. I think using the DTN data forwarding (whether someone will forward data for others) , we can explore the real altruistic/selfish behavior of the people. If we find out that people are very willing to forward data then it is good for networking conference paper, if we find out that people are not really willing to forward data, we can have a paper to Journal of Economic Behavior and Organization or even to Science.  So overall, I believe we are doing very good and important research.


The single most important question in the whole design is how to ensure that the GKs are non-malicious (and not spammers themselves). Obviously, we cannot simply assume they are non-malicious simply based on the fact that they are in the social network. Otherwise we can simply whitelist the whole social network.  
Regarding experiment design, you may want to have a look at these papers. I can also scan a chapter of a book (superfreakonomics) by (http://pricetheory.uchicago.edu/levitt/index.html ) I have to send to you, which introduce the problem very well.  
 
The goal of this thesis is to design and implement a protocol for authenticating that the selected GKs (to vouch for spam free communication outside a user’s social circle) are legitimate and non-malicious GKs.  


== Required Skills==
== Required Skills==
Line 21: Line 18:
* Basic understanding of computer networking
* Basic understanding of computer networking
* Good programming skills
* Good programming skills
== Initial Reading ==
[http://www.nber.org/papers/w15701 So you want to run an experiment, now what? Some Simple Rules of Thumb for Optimal Experimental Design]
[http://www.net.informatik.uni-goettingen.de/people/sufian_hameed Sufian Hameed]
[http://www.net.informatik.uni-goettingen.de/people/sufian_hameed Sufian Hameed]
Retrieved from "https://wiki.net.informatik.uni-goettingen.de/wiki/Evaluation_of_Human_Altruism_with_DTN_based_data_forwarding"