Social Trends Analysis using Efficient Burst Detection in Adaptive Time Windows

Abstract

Real time monitoring of twitter tweet streams for trends has popularity in the last decade. This provide effective information for government, business and other organization to know what happening right now. The task comprise many challenges including the processing of large volume of data in real time and high levels of noise. The main objective of this work is timely detection of bursty trends which have happened recently and discovery of their evolutionary patterns along the timeline. We present burst detection in adaptive time windows. It is the task of finding unexpected change in some quantity in real time tweet stream. Burst is highly depend on the sampled time window size and threshold values. So in this work, we describe how to adjust time windows sizes and threshold values in real time. Our experimental results show both processing time is efficient and effectiveness of our approach.