Building a Personalized eCommerce Agent: Introducing the Thrive.AI app
Share this Session:
  Marc Hadfield   Marc C. Hadfield
Vital AI


Wednesday, June 6, 2012
02:30 PM - 03:00 PM
Level:  Technical - Intermediate

Location:  Plaza B

With the advent of applications such as Apple’s Siri and IBM’s Watson, semantic agents are becoming part of our daily lives, providing new levels of human/computer interaction.

This presentation introduces Thrive.AI, a personalized semantic shopping agent for the IPad, and describes it’s inner workings, which combine a number of semantic and Big Data technologies into a seamless user experience.

At the heart of the Thrive.AI application is an ontology which provides the framework for sharing of RDF-encoded data across all subsystems, including delivery to the IPad.

The components of Thrive.AI include:

• IPad application with Speech Recognition
• Allegrograph Triplestore with Groovy Interface
• Frame Logic Rule & Inference system
• Interest Graph Analysis using Scala
• Natural Language Processing & Sentiment Analysis
• Machine Learning & Hadoop

Together, these components provide Big Data processing power at the fingertips of end users to solve everyday problems, such as improving daily purchasing decisions.

The presentation will discuss data integration and scaling across these various technologies, and describe how practical semantic agents can be successfully created and deployed in many user scenarios, while avoiding potential pitfalls.

Marc Hadfield is the founder of Vital AI, and is an expert in information retrieval, knowledge modeling, and software architecture. Previously Marc was CTO of Inform Technologies, a SaaS company for enterprise publishers, and CTO of Alitora Systems, a SaaS company serving Life Science research in drug-discovery. He has designed and built data management systems for the finance and pharmaceutical industries, as well as for genomic research. He is a graduate of Princeton University with a degree in Computer Science Engineering.

Close Window