作者: Giovanni Stilo , Paola Velardi
DOI: 10.1162/COLI_A_00277
关键词:
摘要: Hashtags are creative labels used in micro-blogs to characterize the topic of a message/discussion. Regardless use for which they were originally intended, hashtags cannot be as means cluster messages with similar content. First, because created spontaneous and highly dynamic way by users multiple languages, same can associated different hashtags, conversely, hashtag may refer topics time periods. Second, contrary common words, disambiguation is complicated fact that no sense catalogs e.g., Wikipedia or WordNet available; and, furthermore, difficult analyze, often consist acronyms, concatenated so forth. A determine meaning has been analyze their context, but, we have just pointed out, variable meanings. In this article, propose temporal clustering algorithm based on idea semantically related synchronous usage patterns.