Main Article Content
Abstract
Twitter is one of the communication channels for spreading breaking news. It is an interesting platform for dissemination of news. The nature and brevity of the tweets are conducive way to share information related to important events. But one of the greatest challenges is to find the number of tweets that characterized as breaking news in the ocean of tweets. A novel method is used for detecting and tracking breaking news from Twitter in real-time. Filtering the stream of incoming tweets and it removes junk tweets by using greedy and text classification algorithm. It Compares the performance of different task and clusters the similar tweets. Finally, dynamic scoring system is used to track the news over a period of time. This method is used to collect, group, track and update the breaking news automatically. This provides a convenient way for people to follow the breaking news and stay connected with real-time updates. The domain-specific Naive Bayes model can capture the specific sentiment expressions in each domain. Two kinds of domain similarity measures are explored, one based on textual content and the other one based on sentiment expressions. In this method an efficient way to accurately categorize trending topics without need of external data, enabling news organizations to discover breaking news in realtime, or to quickly identify viral memes that might enrich marketing decisions, among others.