Reasoning Over Big Data Stores
Share this Session:
  Eric Little   Eric Little
Vice President, Data Science
OSTHUS Inc.
 


 

Wednesday, August 19, 2015
10:00 AM - 10:45 AM

Level:  Technical - Introductory


Current challenges facing the implementation of NoSQL-type databases involve how to use advanced rule-based analytics on large tables and key value stores, where metadata is often sparse. Graph databases or triple stores are great for utilizing one’s metadata, but are often computationally inefficient compared to NoSQL stores. To combat this problem, Modus Operandi will showcase a Predicate Store inside of its MOVIA product that can run advanced, first-order level, logical rule sets and queries against large tables or column stores directly to provide a scalable, rapid and advanced data analytics for cloud applications. This provides graph complexity in terms of content with the performance and scalability of NoSQL data approaches. The system also allows for both statistical algorithms as well as logic-based rule sets to be run concurrently, meaning that a host of parallel analytics can be run at once, providing deep analysis over a multitude of important pattern types.


Eric Little is the Vice President of Data Science for OSTHUS and Zontal Inc., where he is working on the research and development of semantic technologies and their application to analytics, including graph-based computing, data modeling, and advanced computational reasoning within pharmaceutical and biotech verticals. Dr. Little works with pharma and biotech customers to design and implement new and cutting edge technologies for improved data management that can ultimately lead to new types of solutions and specialized analytics-based products.


   
Close Window