Textual Entailment as Syntactic Graph Distance: a rule based and a SVM based approach

Posted in Science on August 19, 2008


Textual Entailment as Syntactic Graph Distance: a rule based and a SVM based approach

Lecture slides:

  • Textual Entailment as Syntactic Graph Distance: a rule based and a SVM based approach
  • Classifying Textual Entailment (TE)
  • Recognizing Textual Entailment (TE)
  • Graph Matching (GM)
  • Textual Entailment as Graph Matching (GM)
  • What’s next
  • Extended Dependency Graph (XDG)
  • GM on XDG: definitions
  • Finding the bijective function and evaluating the measure
  • Constituent Similarity
  • Dependency Similarity
  • Textual Entailment Measure
  • Some more details
  • Estimating Parameters with SVM
  • Feature Spaces
  • Feature Spaces
  • Used Resources
  • Preliminary analysis (Rule-based System)
  • Preliminary analysis (SVM-based system)
  • Out from the Fairy Tale...
  • ... and back to real life!!!!

Author: Fabio Massimo Zanzotto, University Of Milano Bicocca

Watch Video

Tags: Science, Lectures, Computer Science, Machine Learning, VideoLectures.Net, Kernel Methods