论文信息 - The NYU Cold Start System for TAC 2015

The NYU Cold Start System for TAC 2015

The KBP Cold Start task builds a knowledge base from scratch using a given document collection and a predefined schema for the entities and relations. We describe the NYU submission to the TAC 2015 Cold Start track (KB variant). We report the overall architecture, new modules we introduce this year, experimental results, and experiments we conduct on building KBP relation extractors using an active learning tool developed for NLP novices.

Yifan He | Ralph Grishman

[1] Yifan He,et al. New York University 2014 Knowledge Base Population Systems , 2014 .

[2] Christopher D. Manning,et al. SPIED: Stanford Pattern based Information Extraction and Diagnostics , 2014 .

[3] Heng Ji,et al. Seed-Based Event Trigger Labeling: How far can event descriptions get us? , 2015, ACL.

[4] Mihai Surdeanu,et al. A Domain-independent Rule-based Framework for Event Extraction , 2015, ACL.

[5] Frederick Reiss,et al. Profile Extractor Test Extractor Develop Extractor Input Documents Label Text / Clues Task Analysis Rule Development Performance Tuning Delivery Export Extractor , 2012 .

[6] Luis Gravano,et al. Snowball: extracting relations from large plain-text collections , 2000, DL '00.

[7] Ralph Grishman,et al. An Efficient Active Learning Framework for New Relation Types , 2013, IJCNLP.

[8] Ralph Grishman,et al. New York University 2012 System for KBP Slot Filling , 2012, TAC.

[9] Ralph Grishman,et al. Active learning for relation type extension with local and global data views , 2012, CIKM '12.

[10] Christopher D. Manning,et al. Combining Distant and Partial Supervision for Relation Extraction , 2014, EMNLP.

[11] Yifan He,et al. ICE: Rapid Information Extraction Customization for NLP Novices , 2015, HLT-NAACL.

[12] Alan Akbik,et al. Propminer: A Workflow for Interactive Information Extraction and Exploration using Dependency Trees , 2013, ACL.