In this talk Noel will discuss a class of algorithms, streaming algorithms, that allow real-time processing of data, scale extraordinarily well, and are simple to implement. Streaming algorithms were developed to handle data capacities that exceed our ability to store them (e.g. the Large Hadron Collider) but turn out to be a great fit for small teams that want to move quickly on data analysis projects.

In this talk Noel will describe the Bloom filter and Count-Min sketch for counting item occurrence and heavy hitter and quantile algorithms for estimating frequency information. If there is time and interest the talk will discuss lock-free stochastic gradient descent for learning classifiers and recommendation systems

	Tools for Haskell (Game) Programming : 19th Jun 2013
	Machine Learning Hands On with F# Tomas Petricek : 13th Jun 2013
	Simplifying Asynchronous Code with Scala Philipp Haller : 5th Jun 2013
	Deuce is (not yet) Emacs under Clojure Hakan Raberg : 4th Jun 2013
	Functional Programming in the enterprise Jon Pither : 3rd Jun 2013
	F# on iPad and iPhone with Xamarin Studi : 30th May 2013
	F# Type Providers Hands On Tomas Petricek : 23rd May 2013
	Be a better developer : 21st May 2013
	Developing Python with PyDev : 21st May 2013
	Machine Learning with Storm + Redis Yodit Stanton : 7th May 2013
	Cascalog for the 80% of Data Science Bruce Durling : 7th May 2013
	F# embedded in Excel Adam Mlocek : 25th Apr 2013
	F# embedded in Excel part 2 Adam Mlocek : 25th Apr 2013
	Pitfalls Of Using Frameworks Paul Brian : 18th Apr 2013
	Adventures with Cython Emil Vaughan : 18th Apr 2013