Real-time Twitter analysis with Reactive Extensions

on YouTube ↗01h 02m

" summary

Live Twitter streams get replayed from election and Q&A episodes and processed with RxJS to produce tweets‑per‑minute rates (peaks ~800 tpm) using rolling one‑minute windows starting every 30s and word histograms over five‑minute windows via reduce into a hashmap. A dashboard replays recorded streams at 500x speed, spike detection compares current vs. normal (delayed) windows and uses a 15% threshold for spikes to emit triggers and the top topics driving each peak; tuning is required but the pipeline works.

AI-written summary · about

" chapters (7)

1. 0:00Introduction to Rx
2. 2:56Measuring stream activity
3. 6:00Analyzing topic trends
4. 17:25Identifying traffic spikes
5. 35:12Linking spikes to topics
6. 48:33Practical applications of Rx
7. 54:31Q&A and conclusion

" related talks

• 21mReal-Time Stream Processing with Reactive-Extensions (RX)Michael Hladky · jsPoland 2018
• 32mReal-Time Stream Processing with Reactive-ExtensionsMichael Hladky · ngVikings 2018
• 28mReactive all the ThingsBen Lesh, Martin Gontovnikas · ng-conf 2015
• 01h 21mReactive Web ApplicationsRossen Stoyanchev, Stephane Maldini · SpringOne2GX 2015
• 43mConnecting Reactive Applications with Fast Data Using Reactive StreamsLuc Bourlier · Scala Days Berlin 2016
• 01h 01mGet Reactive with RxJSVenkat Subramaniam · NDC London 2017
• 19mReactive Programming with RxPyValery Calderon · PyCon 2018
• 56mReal Time Data PipelinesSamantha Quiñones · PHP UK Conference 2016