National Seminar on Recent Trends in Data Mining |
Foundation of Computer Science USA |
RTDM2016 - Number 1 |
April 2016 |
Authors: Rajani D. Gavali, A.r. Kulkarni |
f9c2e44b-5609-4686-a739-861d72c796cd |
Rajani D. Gavali, A.r. Kulkarni . Topic Detection and Summarization of Events on Social Media Data. National Seminar on Recent Trends in Data Mining. RTDM2016, 1 (April 2016), 3-5.
Millions of Internet users use social sites for sharing and storing data, blogs for connecting with people and sharing information. Twitter is one of the fastest growing social sites. Short-text messages are being posted and shared at a unique rate. Twitter collects millions of tweets, which contain lots of noise and redundancy, un-structured tweets. As redundancy leads to inconsistency and less accurate result, users are unable to understand current topics of discussion. Due to time constraints readers are unable to read each and every tweet, so user requires summary to understand important information on social media. To generate summary for large amount of data, a new summarization method is proposed, namely sequential summarization, which provides a topic detection of social media and generate ordered short sub-summaries for a trending topic in order to convey the important information in few sentences. The system will implement two approaches, stream-based and semantic-based, for detecting the necessary and non-redundant subtopics within trending information.