Get token string from tokenID using Stanford Parser in GATE

Question

I am trying to use some Java RHS to get the string value of dependent tokens using Stanford dependency parser in GATE, and add them as features of a new annotation.

I am having problems targeting just the 'dependencies' feature of the token, and getting the string value from the tokenID.

Using below specifying only 'depdencies' also throws a java null pointer error:

for(Annotation lookupAnn : tokens.inDocumentOrder())
  {
   FeatureMap lookupFeatures  = lookupAnn.getFeatures();
   token = lookupFeatures.get("dependencies").toString();  
  }

I can use below to get all the features of a token,

gate.Utils.inDocumentOrder

but it returns all features, including the dependent tokenID's; i.e:

dependencies = [nsubj(8390), dobj(8394)]

I would like to get just the dependent token's string value from these tokenID's.

Is there any way to access dependent token string value and add them as a feature to the annotation?

Many thanks for your help

dedek · Accepted Answer · 2018-06-07T08:30:05.593

Here is a working JAPE example. It only printns to the GATE's message window (std out), It doesn't create any new annotations with features you asked for. Please finish it yourself...

Stanford_CoreNLP plugin has to be loaded in GATE to make this JAPE file loadable. Otherwise you will get class not found exception for DependencyRelation class.

Imports: {
  import gate.stanford.DependencyRelation;
}

Phase: GetTokenDepsPhase
Input: Token
Options: control = all
Rule: GetTokenDepsRule
(
  {Token}
): token
--> 
:token {
  //note that tokenAnnots contains only a single annotation so the loop could be avoided...
  for (Annotation token : tokenAnnots) {
    Object deps = token.getFeatures().get("dependencies");

    //sometimes the dependencies feature is missing - skip it
    if (deps == null) continue;

    //token.getFeatures().get("string") could be used instead of gate.Utils.stringFor(doc,token)...
    System.out.println("Dependencies for token " + gate.Utils.stringFor(doc, token));

    //the dependencies feature has to be typed to List<DependencyRelation>
    List<DependencyRelation> typedDeps = (List<DependencyRelation>) deps;
    for (DependencyRelation r : typedDeps) {

      //use DependencyRelation.getTargetId() to get the id of the target token
      //use inputAS.get(id) to get the annotation for its id
      Annotation targetToken = inputAS.get(r.getTargetId());

      //use DependencyRelation.getType() to get the dependency type
      System.out.println("  " +r.getType()+ ": " +gate.Utils.stringFor(doc, targetToken));
    }
  }
}

I cannot get this to work... using your complete rule copied/pasted doesn't produce any messages to GATE window (std out). And when I simplify and try the following I get a null pointer exception: for (Annotation token : tokens.inDocumentOrder()) { FeatureMap lookupFeatures = token.getFeatures(); deps = token.getFeatures().get("dependencies").toString(); } I have loaded corenlp plugin and the following imports... any suggestion? Thanks in advance for your help. Imports: { import static gate.Utils.*; import gate.stanford.DependencyRelation; } — FJ1993, Jun 08 '18 at 17:06
@FJ1993 No problem with chat, but I'm not sure if we find a common time slot... My jape file has to be executed on some non-empty GATE document processed by stanford parser. I assume you have one... Or is it resolved already? — dedek, Jun 11 '18 at 15:51

Get token string from tokenID using Stanford Parser in GATE

1 Answers1