Beehive based machine to give snapshot of the ongoing stories on the web

Abstract

In this paper we present an approach, inspired by honey bees, that allows us to take a glance at current events by exploring a portion of the Web and extracting keywords, relevant to current news stories. Not unlike the bees, that cooperate together to retrieve little bits of food, our approach uses agents to select random keywords and carry them from one article to another, landing only on the articles relevant to the keyword. Keywords that best represent multiple articles are selected, while keywords not relevant to articles are subsequently discarded and not explored further. Our results show, that with this approach, it is possible to extract keywords relevant to news stories, without utilizing learning methods, or analysis of a data corpus. © 2013 Springer-Verlag Berlin Heidelberg.