resources.coref.ace-pronouns.jape Maven / Gradle / Ivy

Go to download
Show more of this group Show more artifacts with this name
Show all versions of annie Show documentation
ANNIE is a general purpose information extraction system that provides the building blocks of many other GATE applications.
There is a newer version: 9.1
Show newest version
/*
*  ace-pronouns.jape
*
* Copyright (c) 1998-2004, The University of Sheffield.
*
*  This file is part of GATE (see http://gate.ac.uk/), and is free
*  software, licenced under the GNU Library General Public License,
*  Version 2, June 1991 (in the distribution as file licence.html,
*  and also available at http://gate.ac.uk/gate/licence.html).
*
*  Marin DImitrov, 19/Dec/2001
* 
*  $Id: ace-pronouns.jape 5922 2004-07-21 17:10:11Z akshay $
*/


Phase:	count_pronouns
Input: Token
Options: control = brill

Rule:countPronouns
Priority: 99
(
    (
        {Token.category==PRP,Token.string==I}  |      
        {Token.category==PRP,Token.string==Me}  |      
        {Token.category==PRP,Token.string==me}  |      
        {Token.category==PRP,Token.string==You}  |
        {Token.category==PRP,Token.string==you}  |            
        {Token.category==PRP,Token.string==He} |  
        {Token.category==PRP,Token.string==he}  |
        {Token.category==PRP,Token.string==Him} |  
        {Token.category==PRP,Token.string==him}  |
        {Token.category==PRP,Token.string==She} |  
        {Token.category==PRP,Token.string==she}  |
        {Token.category==PRP,Token.string==Her} |  
        {Token.category==PRP,Token.string==her}  |
        {Token.category==PRP,Token.string==It} |  
        {Token.category==PRP,Token.string==it}  |
        {Token.category==PRP,Token.string==We}  |
        {Token.category==PRP,Token.string==we}    |
        {Token.category==PRP,Token.string==Us}  |
        {Token.category==PRP,Token.string==us}    |
        {Token.category==PRP,Token.string==They} |  
        {Token.category==PRP,Token.string==they}  |
        {Token.category==PRP,Token.string==Them} |  
        {Token.category==PRP,Token.string==them}  |

        {Token.category=="PRP$",Token.string==My}  |                            
        {Token.category=="PRP$",Token.string==my}  |                        
        {Token.category==PRP,Token.string==Mine}  |      
        {Token.category==PRP,Token.string==mine}  |      
        {Token.category=="PRP$",Token.string==Your}  |
        {Token.category=="PRP$",Token.string==your}  |
        {Token.category=="PRP$",Token.string==Yours}  |
        {Token.category=="PRP$",Token.string==yours}  |
        {Token.category=="PRP$",Token.string==His} |  
        {Token.category=="PRP$",Token.string==his}  |
        {Token.category=="PRP$",Token.string==Her} |  
        {Token.category=="PRP$",Token.string==her}  |
        {Token.category==PRP,Token.string==Hers}  |        
        {Token.category==PRP,Token.string==hers}  |                
        {Token.category=="PRP$",Token.string==Its} |  
        {Token.category=="PRP$",Token.string==its}  |
        {Token.category=="PRP$",Token.string==Our}  |
        {Token.category=="PRP$",Token.string==our}   |
        {Token.category==PRP,Token.string==Ours}  |
        {Token.category==PRP,Token.string==ours}    |
        {Token.category=="PRP$",Token.string==Their} |  
        {Token.category=="PRP$",Token.string==their} 
        {Token.category==PRP,Token.string==Theirs} |  
        {Token.category==PRP,Token.string==theirs}  |                
        
        {Token.category==PRP,Token.string==Myself}  |      
        {Token.category==PRP,Token.string==myself}  |                                                                
        {Token.category==PRP,Token.string==Yourself}  |
        {Token.category==PRP,Token.string==yourself}  |                    
        {Token.category==PRP,Token.string==Himself} |  
        {Token.category==PRP,Token.string==himself}  |
        {Token.category==PRP,Token.string==Herself}  |                        
        {Token.category==PRP,Token.string==herself}  |                
        {Token.category==PRP,Token.string==Itself}  |
        {Token.category==PRP,Token.string==itself}  |                
        {Token.category==PRP,Token.string==Oneself}  |
        {Token.category==PRP,Token.string==oneself}  |                

        {Token.category==PRP,Token.string==Ourselves}  |
        {Token.category==PRP,Token.string==ourselves}    |
        {Token.category==PRP,Token.string==Yourselves}  |
        {Token.category==PRP,Token.string==yourselves}    |
        {Token.category==PRP,Token.string==Themselves} |  
        {Token.category==PRP,Token.string==themselves}            
    ) :PRP
)
-->
{
    gate.AnnotationSet pronouns = (gate.AnnotationSet)bindings.get("PRP");
    Annotation prn = (Annotation) pronouns.iterator().next();
    
    String prnType = ((String)prn.getFeatures().get("string")).toLowerCase();
    
    //try to get doc feature    
    Integer cnt = (Integer)doc.getFeatures().get(prnType);

    if (null == cnt) {
        cnt = new Integer(1);
    }
    else {
        cnt = new Integer(cnt.intValue() + 1);
    }
    
    doc.getFeatures().put(prnType,cnt);
}



Rule:countWords
Priority: 11
(
  (
        {Token}         
    ) :word  
)
-->
{
    Integer currCount = (Integer)doc.getFeatures().get("WORDS");

    if (null == currCount) {
        currCount = new Integer(1);
    }
    else {
        currCount = new Integer(currCount.intValue() + 1);
    }
    
    doc.getFeatures().put("WORDS",currCount);
}