Overview
Explore the revolution in text analytics accelerated by Machine Learning advancements in human language technologies. Learn about rapid triage and exploitation of digital media in this 25-minute talk from #HLTCon 2018. Discover how to efficiently process terabytes of text data, moving beyond traditional Boolean and keyword searches to find the most significant documents for human translation and analysis. Gain insights into innovative techniques such as concept search, text embedding, enhanced identity, and clustering to break the "warp barrier" in digital media processing.
Syllabus
Introduction
Background
Park
Parallelcorpora
Bullying indexing
Start with why
Artificial stupidity
Machine Translation
Information Overload
Henry Ford Quote
Warp to Prototype
Prototype
Individual Files
Filtering
Boolean Search
Concept Search
Text Embedding
Enhanced Identity
Clustering
Questions
Taught by
BasisTech