PROJECT TITLE:

TopicSketch Real-time Bursty Topic Detection from Twitter - 2016

ABSTRACT:

Twitter has become one among the largest microblogging platforms for users around the planet to share anything happening around them with friends and beyond. A bursty topic in Twitter is one that triggers a surge of relevant tweets within a brief period of your time, which often reflects important events of mass interest. How to leverage Twitter for early detection of bursty topics has thus become an necessary research problem with immense practical price. Despite the wealth of research work on topic modelling and analysis in Twitter, it remains a challenge to detect bursty topics in real-time. As existing ways can hardly scale to handle the task with the tweet stream in real-time, we have a tendency to propose in this paper ?????????????????????? , a sketch-primarily based topic model together with a group of techniques to achieve real-time detection. We have a tendency to evaluate our solution on a tweet stream with over 30 million tweets. Our experiment results show each efficiency and effectiveness of our approach. Particularly it is conjointly demonstrated that ?????????????????????? on one machine will doubtless handle tons of millions tweets per day, that is on the identical scale of the entire number of daily tweets in Twitter, and gift bursty events in finer-granularity.