High-Throughput Continuous Clustering of Message Streams
Marek Landowski, Guangyu Wu · 2013
Continuous streaming data is commonplace across a wide range of domains and industries, and analysing these often high volume data streams in an on-line, low latency and scalable way can be challeng- ing. We present a continuous clustering system for topic detection and monitoring in high-throughput message streams. A parallel implementa- tion of sequential leader clustering runs on the Storm 1 stream processing