Thanks for attending our meetup in London!
We hoped you enjoyed this meetup as much as we did.
Get the slides of the talk here!
Anomaly Detection in Aviation: A Dataikulx Vertica Analytics Project
First Talk: "Getting the Data: Tracking commercial aircraft in near real-time using a Raspberry Pi, Kafka & Vertical"
In this presentation, Mark Whalley, Systems Engineer at Microfocus, shows how it possible to track the flights of commercial aircraft in near real-time using relatively simple to acquire hardware and software. Automatic Dependent Surveillance Broadcast (ADS-B) data from aircraft transponders is captured and decoded using a Raspberry Pi, delivered through a series of Apache Kafka Topics before being loaded into a Vertica database.
With real-time streaming combined with billions of rows of historical data, we look at how this can be manipulated and used with relative ease using the inbuilt analytics and machine learning capabilities of Vertica for data capture, enrichment, measuring and preparing. Demonstrating examples of simple SQL functions for handling time series data, gap filling and interpolation and outlier detection.
Second Talk: "Putting the Data to Work: Anomaly Detection with Time Series + Production"
Building on Mark's presentation, Silviu Tofan, Data Scientist at Dataiku looks at a practical use case of anomaly detection. Going first through exploratory steps, he walks through the definition of our anomaly detection case, how this is achieved using time series data, as well what are the best way of putting such a model into production.
