Nature-Inspired Dissemination of Information in P2P Networks

Abstract

After having first been used as ameans to publish content, theWeb is now widely used as a social tool for sharing information. It is an easy task to subscribe to a social network, join one of theWeb-based communities according to some per- sonal interests and start to share content with all the people who do the same. It is easy once you solve two basic problems: select the network to join (go to hi5, face- book, myspace,:::? join all of them?) and find/pick up the right communities (i.e., find a strict label to match non-strict centers of interest). An error of appreciation would result in getting too much of useless/non-relevant information. This chapter provides a study on the dissemination of information within groups of people and aimat answering one question: can we find an effortlessway of sharing information on theWeb? Ideally, such a solution would require neither the definition of a profile nor the selection of communities to join. Publishing information should also not be the result of an active decision but be performed in an automatic way. A nature- inspired framework is introduced as an answer to this question. This framework features artificial ants taking care of the dissemination of information items within the network. Centers of interest of the users are reflected by artificial pheromones laid down on connections between peers. Another part of the framework uses those pheromone trails to detect shared interests and creates communities.