• CLARIN-PL Repository Home
  • View Item
  •  
  •   What can you do?
  •   Browse  
    •    All of the Repository  
      •   Issue Date
      •   Authors
      •   Titles
      •   Subjects
      •   Publisher
      •   Language
      •   Type
      •   Rights Label
  •   My Account  
    •    Login via Your home institution
    •    Register
  •   Statistics  
    •    StatisticsBETA
  •   General Information  
    •    Deposit
    •    Cite
    •    Submission Lifecycle
    •    FAQ
    •    About and Policies
    •    Help Desk
 
 

NamEnSim

 
CLARIN-PL
  Authors
Kocoń, Jan
 Project URL
http://nlp.pwr.wroc.pl/pl/narzedzia-i-zasoby/namensim
 Date issued
2012-04-04
 Type
toolService
 Language(s)
Polish
 Description
NamEnSim (Named Entity Similarity Function) is a tool to evaluate the value of the word pair similarity, using heterogeneous named entity similarity function. This tool allows to get the value of similarity between unknown word and items from gazetteers (e.g. person first names, person last names, country names). NamEnSim can also extract multi-word named entities from the source text using similarity function. This tool supports named entities recognition task (NER), e.g. Liner2. In the named entities "base - form" matching task NamEnSim achieves the accuracy of 81-96% for the similarity function based on logistic regresion and 84-98% for the similarity function based on inflection rules (results depend on the named entity category).
 Publisher
Wrocław University of Technology
 Collection(s)
CLARIN-PL
Show full item record
 
 
  • © 2024 CLARIN-PL. All Rights Reserved.
  • Base on DSpace modified by UFAL MFF UK and CLARIN-PL
  • Privacy policy | Licenses