Claudia Perlich discusses privacy-preserving representations, robust high-dimensional modeling, large-scale automated learning systems, transfer learning, and fraud detection.
Paco Nathan keynotes on how Spark fits into the big data landscape, describing what other systems work with Spark, and explaining why Spark is needed in the future.
Jim Scott keynotes on the history of Hadoop, the difficulties that this technology has gone through, exploring the reasons why enterprises need to evaluate their targets and prepare for the future.
John Zamierowski discusses the business benefits of big data coming from the Internet of Everything, focusing on the "Why" and "How" of big data and current developments in sensor technology.
Stefan Edlich discusses big data systems -Spanner, Presto- and the future of data persistence, data analytics, data formats and of NoSQL/NewSQL in general.
John Canfield discusses the changing payment ecosystem, innovations in mining and organizing unstructured data from many sources, and approaches to deciding for loss minimization and user experience.
Josh Long introduces some of the latest Spring features supporting HATEOAS-compliant and OAuth-secured REST services, NoSQL and Big Data, Websockets, OAuth, open-web security and mobile.
The authors present design patterns and use cases of capital market firms that are incorporating big data technologies into their credit risk analysis, price discovery or sentiment analysis software.
Nellwyn Thomas discusses how Etsy is using A/B testing, how Etsy's data-driven culture has evolved over time and how continuous delivery and big data can coexist.
Shawn Gandhi overviews real-time processing use cases, and how developers are using AWS Kinesis to shift from a traditional batch-oriented approach to a continual real-time data processing model.
Randy Shoup describes KIXEYE's analytics infrastructure from Kafka queues through Hadoop 2 to Hive and Redshift, built for flexibility, experimentation, iteration, testability, and reliability.
Marcel Kornacker presents a case study of an EDW built on Impala running on 45 nodes, reducing processing time from hours to seconds and consolidating multiple data sets into one single view.