Cascaded Information Synthesis for Timeline Construction

Abstract

This paper presents a method for incrementally constructing CEO succession timelines via a cascade of minimally supervised information synthesis components. Results from multiple documents are combined using a CRF fusion method. The system demonstrates that with minimal training on the target domain, the presence of redundant information allows for the synthesis of networks of interrelated facts from text.

Open PDF

Document Details

Document Type
Technical Report
Publication Date
Jan 01, 2006
Accession Number
ADA448478

Entities

People

  • Gideon Mann

Organizations

  • University of Massachusetts Amherst

Tags

Communities of Interest

  • Biomedical

DTIC Thesaurus Topics

  • Artificial Intelligence
  • Artificial Intelligence Software
  • Automata Theory
  • Computational Science
  • Computer Languages
  • Computer Science
  • Construction
  • Data Mining
  • Databases
  • Information Retrieval
  • Information Science
  • Language
  • Machine Learning
  • Models
  • Named Entity Recognition
  • Natural Language Processing
  • Probabilistic Models

Fields of Study

  • Computer science

Readers

  • Applied Combinatorial Optimization and Logic Circuit Design.
  • Economics
  • Neural Network Machine Learning.