DISCOVERING EMERGING TOPICS IN SOCIAL STREAMS VIA LINK ANOMALY DETECTION

Abstract

Discovery of emerging topics is now
getting converted interest motivated by the
rapid growth of social networks. We focus
on surfacing of topics signaled by social
aspect of these networks. Specifically, we
focus on mentions of users—link among
users that are generated energetically
through replies, mentions, and retweets. We
propose a probability model of the
mentioning performance of a social network
user, and propose to detect the emergence of
a new topic. We demonstrate our method in
numerous real data sets we gathered from
Twitter. The experiments show that the
proposed mention – anomaly - based
approaches can also be detected by new
topics at least as early as text-anomalybased
approaches, and in some cases
greatly before when the topic is badly
identified by the textual contents in posts