Research and implementation of event extraction from twitter using LDA and scoring function

Abstract

With the fast growth of social media, interest is increasing in detecting popular events from tweets. Event extraction is a work which identifies events from tweets or database of tweets. Each and every day, hundreds of megabytes of current stories are being added into the news archives of the major news agencies, containing much important and interesting news. Aim of this event extraction strategy is to extract and retrieve major life events from twitter data. Example of events extraction include seminar presentation, Job opening, Admission in Top universities, new technology etc. The role of this extraction is to collect major life events in the form of retrievable entries that include structured data about major life event name, location and time. Most of previous research on event extraction was mainly on textual level extraction such as News, medical systems, text summarization, whereas less work has been done on event extraction from noisy text such as tweets. For instance, tweets are short and self-contained which make them lack useful information. The target of this research is to develop algorithm and methodology that extract and efficiently conclude major life events extracted from social media.