Lars Backstrom
homepage:http://www.cs.cornell.edu/~lars/
search externally:   Google Scholar,   Springer,   CiteSeer,   Microsoft Academic Search,   Scirus ,   DBlife

Description

Lars Backstrom is a Data Scientist at Facebook, where he works on data analysis and machine learning systems. Since joining Facebook, he has made significant improvements to the friend suggestion ranking system. He is responsible for the backend system which does feature extraction and ranking of the friends of friends of over 500 million users (times an average of 40,000 friends of friends means 20 trillion rankings). Before joining Facebook in 2009, Lars was a Ph.D. student at Cornell University, where he developed algorithms for data mining in social networks, query logs, and server logs.


Lectures:

invited talk
flag Dealing with structured and unstructured data at Facebook
as author at  8th Extended Semantic Web Conference (ESWC) 2011,
together with: Marko Grobelnik (introducer),
14094 views
  keynote
flag Serving a Billion Personalized News Feeds
as author at  KDD 2016 Workshops,
2447 views
lecture
flag Optimizing Web Traffic via the Media Scheduling Problem
as author at  Research Track,
3656 views