NIST TAC Knowledge Base Population (KBP2016) Entity Discovery and Linking Track

Overview

The Entity Discovery and Linking (EDL) track at NIST TAC-KBP2016 aims to extract named entity mentions from a source collection of textual documents in multiple languages (English, Chinese and Spanish), and link them to an existing Knowledge Base (KB).  An EDL system is also required to cluster mentions for those NIL entities that don't have corresponding KB entries.

What's New

  • Combine with tri-lingual slot filling to form up an end-to-end tri-lingual KBP task

  • Target at a larger scale data processing, by increasing the size of source collections from 500 documents to 90,000 documents

    • 500 “core” documents will be manually annotated for evaluation

  • Specific, individual nominal mentions will be expanded to all entity types and all languages

Updates

  • 3/4/2016: KBP2016 Registration is open.
  • 3/7/2016: Task Spec is up.