edu.stanford.nlp.dcoref

Class SingletonPredictor

java.lang.Object

edu.stanford.nlp.dcoref.SingletonPredictor

public class SingletonPredictor
extends java.lang.Object

Train the singleton predictor using a logistic regression classifier as described in Recasens, de Marneffe and Potts, NAACL 2013 Label 0 = Singleton mention. Label 1 = Coreferent mention. This is an example of the properties file for this class: # This is set to true so that gold POS, gold parses and gold NEs are used. dcoref.replicate.conll = true # Path to the directory containing the CoNLL training files. dcoref.conll2011 = /path/to/conll-2012/v4/data/train/data/english/annotations/ # Output file where the serialized model is saved. singleton.predictor.output = /path/to/predictor.output

Author:

Marta Recasens, Marie-Catherine de Marneffe

Method Summary

Modifier and Type	Method and Description
static void	main(java.lang.String[] args)
static LogisticClassifier<java.lang.String,java.lang.String>	train(GeneralDataset<java.lang.String,java.lang.String> pDataset) Train the singleton predictor using a logistic regression classifier.

Methods inherited from class java.lang.Object

clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

Method Detail

train

public static LogisticClassifier<java.lang.String,java.lang.String> train(GeneralDataset<java.lang.String,java.lang.String> pDataset)

Train the singleton predictor using a logistic regression classifier.

Parameters:

pDataset - Dataset of features

Returns:

Singleton predictor

main

public static void main(java.lang.String[] args)
                 throws java.lang.Exception

Throws:

java.lang.Exception