BEGIN:VCALENDAR
PRODID:-//hacksw/handcal//NONSGML v1.0//EN
METHOD:PUBLISH
BEGIN:VEVENT
DESCRIPTION:Entity Extraction has become one of the key tools of the semantic web. Entity Extraction is a process of identifying key entities such as people, organizations, products, places and events in unstructured text documents and annotating these entities with precise semantic markup for future use. nThis presentation will review the entity extraction process, compare architectures and patterns, and compare how various open source and commercial tools have recently evolved to meet the needs of the semantic web. We will also include demonstrations of entity extraction.n
n
What is entity extraction, and why is it so challenging?
n
What are the benefits of using annotated documents? (search, re-purposing, time line views, geo-mashups etc.)
n
Architecture and patterns: linguistic pipelines, annotations, enrichment, POS tagging/RDF encoding
n
Standards: XML, RDFa, Microformats, TEI, DITA etc.
n
Are entity extraction web services right for you?
n
Linking to domain specific controlled vocabularies: products, terms, people, places
n
Case studies
n
DTSTART:20100625T130000
SUMMARY:Overview of Entity Extraction
DTEND:20100625T162959
LOCATION:Yosemite B
END:VEVENT
END:VCALENDAR