H2O: Scaling Big Data Algorithms
Today we will hear from SriSatish Ambati, founder of 0xdata (pronounced 'hexadata'), which makes H2O, an opensource prediction and math engine for big data. He will be giving a talk on scaling GLM, Random Forest, and other popular big data algorithms such as the AllState Kaggle dataset.
This talk was recorded at the SF Data Mining meetup at Trulia.
Bio: Before founding 0xdata, SriSatish Ambati was the chief tinkerer of Java and Data stacks for the Cassandra Company, DataStax (formerly Riptano) & before that for 7 years at the JVM company, Azul Systems. Over the course of a decade or so, he has profiled, scaled and supported hundreds of in-production and in-development Java, enterprise and database stacks for performance. He uses these experiences to distil system trends and turbulences. SriSatish’s talks have been well received at conferences & tech fairs including JavaOne, JBossWorld & ApacheCon.
You can find more talks on big data here.