作者: Joana Costa , Catarina Silva , Mário Antunes , Bernardete Ribeiro
DOI: 10.1007/978-3-642-37213-1_24
关键词:
摘要: Given the wide spread of social networks, research efforts to retrieve information using tagging from networks communications have increased. In particular, in Twitter network, hashtags are widely used define a shared context for events or topics. While this is common practice often freely introduced by user become easily biased. paper, we propose deal with bias defining semantic meta-hashtags clustering similar messages improve classification. First, use user-defined as message class labels. Then, apply meta-hashtag approach boost performance