Fast, Scalable, and Context-Sensitive Detection of Trending Topics in Microblog Post Streams

Abstract

Social networks, such as Twitter, can quickly and broadly disseminate news and memes across both real-world events and cultural trends. Such networks are often the best sources of up-to-the-minute information, and are therefore of considerable commercial and consumer interest. The trending topics that appear first on these networks represent an answer to the age-old query “what are people talking about?” Given the incredible volume of posts (on the order of 45,000 or more per minute), and the vast number of stories about which users are posting at any given time, it is a formidable problem to extract trending stories in real time. In this article, we describe a method and implementation for extracting trending topics from a high-velocity real-time stream of microblog posts. We describe our approach and implementation, and a set of experimental results that show that our system can accurately find “hot” stories from high-rate Twitter-scale text streams.

GARNET: A holistic system approach for trending queries in microblogs

“…Regarding query types, existing techniques address either frequent or trending items discovery, based on the supported application. Event detection applications [1], [3], [9], [14], [16], [20], [21], [29], however, are more focused on grouping several trending keywords together to report an event rather than focusing on the scalability and performance of retrieving trending keywords. Thus, these applications are orthogonal to our work in a way that GARNET can use one of these techniques in order to report an event based on the returned trending keywords.…”

“…7. Frequent/Trending Items in Microblogs: Event Detection [1], [3], [9], [14], [16], [20], [21], [29]; GeoScope [4]; AFIA [24] keywords at different locations in recent time interval. However, AFIA is only able to answer frequent keywords queries and is not geared towards answering trending queries.…”

Measuring the Semantic Uncertainty of News Events for Evolution Potential Estimation

“…Due to the great pervasive usage and real-time nature [Sakaki et al 2010] of the social network, it becomes the best way or one of the best ways to disseminate and discuss news. The detection methods have been proposed based on the social network, but different factors are considered, like geographic information [Yuan et al 2013], named entities in the texts [Vavliakis et al 2013], word clusters in the texts [Pervin et al 2013], and users' anomaly behaviors [Takahashi et al 2014]. A keyword network is adopted by Zhou and Chen [2014] that is similar to our method, but they only use the community detection on the constructed keyword network to detect events.…”

Stakeholder Analyses of Firm-Related Web Forums

“…The sliding window advances with time and older discussions are excluded, allowing recent participation in the forum to influence the stakeholder analysis. A sliding window approach has been similarly applied in temporal analyses of other social media, such as in Twitter trending topic detection [Pervin et al 2013]. The fading window approach diminishes the influence of discussions over time by weighting their contributions to the stakeholder feature representation based on their age using a decay function.…”