Document Detail


An Efficient Framework for Constructing Generalized Locally-Induced Text Metrics.
MedLine Citation:
PMID:  22278678     Owner:  NLM     Status:  Publisher    
Abstract/OtherAbstract:
In this paper, we propose a new framework for constructing text metrics which can be used to compare and support inferences among terms and sets of terms. Our metric is derived from data-driven kernels on graphs that let us capture global relations among terms and sets of terms, regardless of their complexity and size. To compute the metric efficiently for any two subsets of terms, we develop an approximation technique that relies on the precompiled term-term similarities. To scale-up the approach to problems with huge number of terms, we develop and experiment with a solution that sub-samples the term space. We demonstrate the benefits of the whole framework on two text inference tasks: prediction of terms in the article from its abstract and query expansion in information retrieval.
Authors:
Saeed Amizadeh; Shuguang Wang; Milos Hauskrecht
Related Documents :
22379728 - Validity evidence based on internal structure of scores on the spanish version of the s...
1351308 - Estimation of time since exposure for a prevalent cohort.
22023198 - Multistability and perceptual inference.
22763388 - Function-valued adaptive dynamics and optimal control theory.
12221598 - Multi-component metabolic classification of commercial feverfew preparations via high-f...
25301508 - Simulated annealing based hybrid forecast for improving daily municipal solid waste gen...
Publication Detail:
Type:  JOURNAL ARTICLE    
Journal Detail:
Title:  IJCAI : proceedings of the conference / sponsored by the International Joint Conferences on Artificial Intelligence     Volume:  -     ISSN:  1045-0823     ISO Abbreviation:  -     Publication Date:  2011  
Date Detail:
Created Date:  2012-1-26     Completed Date:  -     Revised Date:  -    
Medline Journal Info:
Nlm Unique ID:  101523961     Medline TA:  IJCAI (U S)     Country:  -    
Other Details:
Languages:  ENG     Pagination:  1159-1164     Citation Subset:  -    
Affiliation:
Intelligent Systems Program, University of Pittsburgh.
Export Citation:
APA/MLA Format     Download EndNote     Download BibTex
MeSH Terms
Descriptor/Qualifier:
Grant Support
ID/Acronym/Agency:
R01 GM088224-02//NIGMS NIH HHS; R01 LM010019-02//NLM NIH HHS

From MEDLINE®/PubMed®, a database of the U.S. National Library of Medicine


Previous Document:  Vegetation dynamics, and land use and land cover change in the Bale Mountains, Ethiopia.
Next Document:  Aryne [3 + 2] cycloaddition with N-sulfonylpyridinium imides and in situ generated N-sulfonylisoquin...