Header menu link for other important links
X
Semantic filtering and event extraction of twitter streams through RDF and SPARQL
Published in Machine Intelligence Research (MIR) Labs
2018
Volume: 10
   
Pages: 208 - 218
Abstract
For long, there has been a huge demand to develop an efficient mechanism to effectively search and extract much needed information from the social web. Manual annotation is effectively possible in information retrieval for limited number of documents, but it is impractical for large accumulation of document retrieval particularly from social media. The principle objective of this research is to comprehend the emergency news by semantically extracting entities and its relations in the posted user generated content. In order to achieve this, the two most sought process of this enablement is that key event identification of every news items and semantic element extraction from those items. These two processes paves the way for building an effective knowledge base as well as a semantic retrieval engines to augment the event level semantic filtration of news items. It has been well observed and noted at many instances that social media's prevalence is a major source of collective formation of large public opinions. Nevertheless, deriving much needed information from it is very useful because of the fact that it is fresh and above all, no one has mediated its content. It is not the news that we have pondered for, but the views of millions of people on the particular events. Hence, this research has turned to the new dimension of identifying the events which are deemed important for many levels of processes. We discover the events by clustering similar information from mainstream social media and categorize the events semantically to enrich the whole process to obtain the much sought discovery of hidden information. © MIR Labs.
About the journal
JournalInternational Journal of Computer Information Systems and Industrial Management Applications
PublisherMachine Intelligence Research (MIR) Labs
ISSN21507988
Open AccessNo