com.rabidgremlin.mutters.opennlp.intent

Class OpenNLPIntentMatcher

java.lang.Object

com.rabidgremlin.mutters.core.ml.AbstractMachineLearningIntentMatcher

com.rabidgremlin.mutters.opennlp.intent.OpenNLPIntentMatcher

All Implemented Interfaces:

com.rabidgremlin.mutters.core.IntentMatcher

public class OpenNLPIntentMatcher
extends com.rabidgremlin.mutters.core.ml.AbstractMachineLearningIntentMatcher

Intent matcher that uses OpenNLP’s document classifier.

Field Summary

Fields inherited from class com.rabidgremlin.mutters.core.ml.AbstractMachineLearningIntentMatcher

DEBUG_MATCHING_SCORES, MAYBE_INTENT_PREFIX

Constructor Summary

Constructors

Constructor and Description
OpenNLPIntentMatcher(java.lang.String intentModel, com.rabidgremlin.mutters.core.Tokenizer tokenizer, com.rabidgremlin.mutters.core.SlotMatcher slotMatcher) Constructor.
OpenNLPIntentMatcher(java.lang.String intentModel, com.rabidgremlin.mutters.core.Tokenizer tokenizer, com.rabidgremlin.mutters.core.SlotMatcher slotMatcher, float minMatchScore, float maybeMatchScore) Constructor.
OpenNLPIntentMatcher(java.net.URL intentModelUrl, com.rabidgremlin.mutters.core.Tokenizer tokenizer, com.rabidgremlin.mutters.core.SlotMatcher slotMatcher, float minMatchScore, float maybeMatchScore) Constructor.

Method Summary

Modifier and Type	Method and Description
protected java.util.SortedMap<java.lang.Double,java.util.SortedSet<java.lang.String>>	generateSortedScoreMap(java.lang.String[] utteranceTokens)

Methods inherited from class com.rabidgremlin.mutters.core.ml.AbstractMachineLearningIntentMatcher

addIntent, getIntents, match

Methods inherited from class java.lang.Object

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

Constructor Detail

OpenNLPIntentMatcher

public OpenNLPIntentMatcher(java.lang.String intentModel,
                            com.rabidgremlin.mutters.core.Tokenizer tokenizer,
                            com.rabidgremlin.mutters.core.SlotMatcher slotMatcher)

Constructor. Sets up the matcher to use the specified model (on the classpath) and specified tokenizer. Defaults to min score match of MIN_MATCH_SCORE and no maybe intent matching.

Parameters:

intentModel - The name of the document categoriser model file to use. This file must be on the classpath.

tokenizer - The tokenizer to use when tokenizing an utterance.

slotMatcher - The slot matcher to use to extract slots from the utterance.

OpenNLPIntentMatcher

public OpenNLPIntentMatcher(java.lang.String intentModel,
                            com.rabidgremlin.mutters.core.Tokenizer tokenizer,
                            com.rabidgremlin.mutters.core.SlotMatcher slotMatcher,
                            float minMatchScore,
                            float maybeMatchScore)

Constructor. Sets up the matcher to use the specified model (on the classpath) and specifies the minimum and maybe match scores.

Parameters:

intentModel - The name of the document categoriser model file to use. This file must be on the classpath.

minMatchScore - The minimum match score for an intent match to be considered good.

maybeMatchScore - The maybe match score. Use -1 to disable maybe matching.

tokenizer - The tokenizer to use when tokenizing an utterance.

slotMatcher - The slot matcher to use to extract slots from the utterance.

OpenNLPIntentMatcher

public OpenNLPIntentMatcher(java.net.URL intentModelUrl,
                            com.rabidgremlin.mutters.core.Tokenizer tokenizer,
                            com.rabidgremlin.mutters.core.SlotMatcher slotMatcher,
                            float minMatchScore,
                            float maybeMatchScore)

Constructor. Sets up the matcher to use the specified model (via a URL) and specifies the minimum and maybe match score.

Parameters:

intentModelUrl - A URL pointing at the document categoriser model file to load.

minMatchScore - The minimum match score for an intent match to be considered good.

maybeMatchScore - The maybe match score. Use -1 to disable maybe matching.

tokenizer - The tokenizer to use when tokenizing an utterance.

slotMatcher - The slot matcher to use to extract slots from the utterance.

Method Detail

generateSortedScoreMap

protected java.util.SortedMap<java.lang.Double,java.util.SortedSet<java.lang.String>> generateSortedScoreMap(java.lang.String[] utteranceTokens)

Specified by:

generateSortedScoreMap in class com.rabidgremlin.mutters.core.ml.AbstractMachineLearningIntentMatcher