Data Ninja Services for Deep Content Understanding
Share this Session:
  Trung Diep   Trung Diep
DOCOMO Innovations, Inc.


Tuesday, August 18, 2015
04:00 PM - 04:30 PM

Level:  Product/Service Offering

Data Ninja is a suite of services for deep text analytics, knowledge graph and semantic information retrieval aimed at organizations of all sizes and levels of expertise. Its main goal is to provide content understanding based on advanced machine learning algorithms offered as a service at affordable price point.

By using the Smart Data service in conjunction with the Smart Content service, developers can detect topics in text as well as be able to navigate the categories and concepts from our knowledge graph for advanced application development.

In this talk, we will demonstrate few capabilities of our SmartContent and SmartData services:

  • detection of concepts, categories and entities in free text
  • detection of sentiment attached to entities and sections of text
  • retrieval of related concepts from a concept graph
  • access to up-to-date concept visitation statistics and concept rank
  • advanced inference in knowledge graphs such as induction and group similarity

We will outline a couple of real-world use cases we encountered so far with our customers.

Trung Diep currently directs the engineering team responsible for delivering the quality and performance of the Data Ninja services at DOCOMO Innovations. Prior to joining Docomo Innovations, Trung has previously worked at Intel, Mercury Interactive, Rambus, and Broadcom, where he was an Associate Technical Director. He received his B.S. and B.A. degrees in Electrical Engineering and Computer Science, respectively, from Rice University and M.S. and Ph.D. degrees in Computer Engineering from Carnegie Mellon University. His research interests span from hardware, such as processor performance modeling and simulation, to software, such as cloud computing technologies, and particularly on the meta layer in which hardware interacts with software. Trung has been granted more than 10 patents covering the areas of branch prediction, multicore arbitration and scheduling, user-level threading, cache memory, memory wear-leveling, and memory encryption.

Close Window