Every week we are introducing new speakers which will be on stage at #bbuzz 2015. Thanks to our program committee we can present part of our new eclectic program. Presentations range from beginner friendly introductions on hot data analysis topics to in-depth technical presentations about scalable architectures. The conference presents more than 50 talks by international speakers specific to the three tags "search", "store" and "scale".
Model as code – How to automate the deployment of R models in production environment
Isabelle is a Data Scientist at Quantmetry, a French consulting start-up ealing with Big Data and Data Science. She has worked in several projects in Machine Learning, where she dealt with outlier detection, community detection in graphs, data visualization or production stakes. Matthieu joined Quantmetry in 2014, where he works as a consultant.
He worked on scalable recommendation systems using Apache Hadoop, Mahout and is now happily embarking on using Spark.
Data Science has enabled companies to establish predictive models about sales, human resources, enhance their customer knowledge and so much more. But what is the afterlife of these models ? Are they doomed to fade away eventually? This talk will present several techniques to automate the deployment of any R model in two complementary production environments : a big Data cluster and a web service allowing for easy reusability
Online learning, Vowpal Wabbit and Hadoop
Scale - 40 min
Heloise is a Senior Data Scientist at Quantmetry, a consulting company in Data Science. Data driven missions in various companies (banking, insurance, industry, energy) are lead by her, whilst she is in charge of the R&D projects at Quantmetry.
Online learning has recently caught a lot of attention. It allows to process massive data, using up a low amount of memory and limited CPU ressources. Heloise will share the the recent experiences Quantmetry´s team has made with online learning and on parallelisation of Vowpal Wabbit on Hadoop in particular. Her talk will shed light on how to run Vowpal Wabbit on Hadoop.