a confidence model for syntactically motivated entailment proofs
Download
Skip this Video
Download Presentation
A Confidence Model for Syntactically-Motivated Entailment Proofs

Loading in 2 Seconds...

play fullscreen
1 / 19

A Confidence Model for Syntactically-Motivated Entailment Proofs - PowerPoint PPT Presentation


  • 83 Views
  • Uploaded on

A Confidence Model for Syntactically-Motivated Entailment Proofs. Asher Stern & Ido Dagan ISCOL June 2011, Israel. Recognizing Textual Entailment (RTE). Given a text, T , and a hypothesis, H Does T entail H. Example. T : An explosion caused by gas took place at a Taba hotel

loader
I am the owner, or an agent authorized to act on behalf of the owner, of the copyrighted work described.
capcha
Download Presentation

PowerPoint Slideshow about ' A Confidence Model for Syntactically-Motivated Entailment Proofs' - nijole


An Image/Link below is provided (as is) to download presentation

Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author.While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server.


- - - - - - - - - - - - - - - - - - - - - - - - - - E N D - - - - - - - - - - - - - - - - - - - - - - - - - -
Presentation Transcript
a confidence model for syntactically motivated entailment proofs

A Confidence Model for Syntactically-Motivated Entailment Proofs

Asher Stern & Ido Dagan

ISCOL

June 2011, Israel

recognizing textual entailment rte
Recognizing Textual Entailment (RTE)
  • Given a text, T, and a hypothesis, H
  • Does T entail H

Example

  • T: An explosion caused by gas took place at a Taba hotel
  • H: A blast occurred at a hotel in Taba.
proof over p arse t rees
Proof Over Parse Trees

T = T0→ T1→ T2→ ... →Tn = H

bar ilan proof system entailment rules
Bar Ilan Proof System - Entailment Rules

Generic Syntactic

Lexical Syntactic

Lexical

explosion

blast

bar ilan proof system
Bar Ilan Proof System

H: A blast occurred at a hotel in Taba.

An explosion caused by gas took place at a Taba hotel

A blast caused by gas took place at a Taba hotel

A blast took place at a Taba hotel

A blast occurred at a Taba hotel

A blast occurred at a hotel in Taba.

Lexical

Lexical syntactic

Syntactic

tree edit distance
Tree-Edit-Distance

Insurgents attacked soldiers -> Soldiers were attacked by insurgents

proof over parse trees
Proof over parse trees

Which steps?

How to classify?

Decide “yes” if and only if a proof was found

Almost always “no”

Cannot handle knowledge inaccuracies

Estimate a confidence to the proof correctness

  • Tree-Edits
    • Regular or custom
  • Entailment Rules
proof systems
Proof systems

TED based

Entailment Rules based

Linguistically motivated

Rich knowledge

No estimation of proof correctness

Incomplete proofs

Mixed system with ad-hoc approximate match criteria

  • Estimate the cost of a proof
  • Complete proofs
  • Arbitrary operations
  • Limited knowledge

Our System

  • The benefits of both worlds, and more!
    • Linguistically motivated complete proofs
    • Confidence model
our method
Our Method
  • Complete proofs
    • On the fly operations
  • Cost model
  • Learning model parameters
on the fly operations
On the fly Operations
  • “On the fly” operations
    • Insert node on the fly
    • Move node / move sub-tree on the fly
    • Flip part of speech
    • Etc.
  • More syntactically motivated than Tree Edits
  • Not justified, but:
  • Their impact on the proof correctness can be estimated by the cost model.
cost model
Cost Model

The Idea:

Represent the proof as a feature-vector

Use the vector in a learning algorithm

cost model1
Cost Model
  • Represent a proof as F(P) = (F1, F2 … FD)
  • Define weight vector w=(w1,w2,…,wD)
  • Define proof cost
  • Classify a proof
    • b is a threshold
  • Learn the parameters (w,b)
search algorithm
Search Algorithm
  • Need to find the “best” proof
  • “Best Proof” = proof with lowest cost
    • Assuming a weight vector is given
  • Search space is exponential
      • pruning
parameter estimation
Parameter Estimation
  • Goal: find good weight vector and threshold (w,b)
  • Use a standard machine learning algorithm (logistic regression or linear SVM)
  • But: Training samples are not given as feature vectors
    • Learning algorithm requires training samples
    • Training samples construction requires weight vector
    • Learning weight vector done by learning algorithm
  • Iterative learning
parameter estimation2
Parameter Estimation
  • Start with w0, a reasonable guess for weight vector
  • i=0
  • Repeat until convergence
    • Find the best proofs and construct vectors, using wi
    • Use a linear ML algorithm to find a new weight vector, wi+1
    • i = i+1
conclusions
Conclusions
  • Linguistically motivated proofs
    • Complete proofs
  • Cost model
    • Estimation of proof correctness
  • Search best proof
  • Learning parameters
  • Results
    • Reasonable behavior of learning scheme
ad