Survey of Techniques for Text Analytics in Unstructured, Semi-Structured, and Structured Data

The presentation provides an architectural overview on how to build infrastructure for various applications that have to search, categorize, manipulate, and draw inferences from a mixture of data types including the spectrum of unstructured documents to semi-structured data that has some structure around textual blocks to fully structured data stored in databases.

The survey includes:

Evaluation of architectural techniques and models for building or evaluating an infrastructure
Tool-sets, algorithms, and technologies that may be appropriate for different types of problems
Discussion of some of the problem spaces (eDiscovery, Risk Management, Journalism, Medical Records Management, etc.) that can leverage these techniques

Mr. B'Far is currently Senior Director of Software Development for Oracle GRC products. GRC Control suite (one of Oracle GRC products) is designed to search and find various configuration, authorization, and transactional violations. It heavily leverages A.I. and Semantic Web technologies (OWL, RDF, etc.). Prior to Oracle, he was at Google and before his term at Google, he the CTO at Voice Genesis Inc. and an Engineering Manager at eBuilt. Previously, his consulting carrier included clients such as First American Title, CircleUp, Rule Space Inc., 8e6 Technologies, Datallegro (acquired by MS), and others a software architect. He participates in W3C's LinkedData and Provenance WG's.