The paper examines how mobile systems can exploit people's social relations to make more informed decisions. This can lead to substantial performance gains and higher query hit rates. In order to do this the authors divided the classified social interactions in two different categories: friends and strangers. (people you meet often for longer periods of time vs. people you meet sporadically for short period of time) Afterwards the authors investigated in the potential of incorporating social information in mobile systems. This was done by using a DTN routing protocol that avoided forwarding to strangers which worked very well when routing between friends. The authors found that in the friend network, there was a high chance of clustering.
It was also found that Another experiment was also conducted to see if a firewall that discriminates between friends and strangers slow down the propagation of a virus in a mobile network. (this was found to be true) It was also found that reliance on strangers are still necessary to allow mobile P2P file-sharing systems to satisfy the user's requests.
No comments:
Post a Comment