com.hankcs.hanlp.seg

Class WordBasedSegment

java.lang.Object

com.hankcs.hanlp.seg.Segment

com.hankcs.hanlp.seg.WordBasedSegment

Direct Known Subclasses:

DijkstraSegment, NShortSegment, ViterbiSegment

public abstract class WordBasedSegment
extends Segment

基于词语NGram模型的分词器基类

Author:

hankcs

Field Summary

Fields inherited from class com.hankcs.hanlp.seg.Segment

config, customDictionary

Constructor Summary

Constructors

Constructor and Description
WordBasedSegment()

Method Summary

Modifier and Type	Method and Description
protected static List<Term>	convert(List<Vertex> vertexList) 将一条路径转为最终结果
protected List<Term>	decorateResultForIndexMode(List<Vertex> vertexList, WordNet wordNetAll) 为了索引模式修饰结果
protected static void	fixResultByRule(List<Vertex> linkedArray) 通过规则修正一些结果
protected static Graph	generateBiGraph(WordNet wordNet) 生成二元词图
protected static void	generateWord(List<Vertex> linkedArray, WordNet wordNetOptimum) 对粗分结果执行一些规则上的合并拆分等等，同时合成新词网
protected void	generateWordNet(WordNet wordNetStorage) 生成一元词网
protected static void	speechTagging(List<Vertex> vertexList) 词性标注

Methods inherited from class com.hankcs.hanlp.seg.Segment

atomSegment, combineByCustomDictionary, combineByCustomDictionary, combineByCustomDictionary, combineByCustomDictionary, convert, enableAllNamedEntityRecognize, enableCustomDictionary, enableCustomDictionary, enableCustomDictionaryForcing, enableIndexMode, enableIndexMode, enableJapaneseNameRecognize, enableMultithreading, enableMultithreading, enableNameRecognize, enableNumberQuantifierRecognize, enableOffset, enableOrganizationRecognize, enablePartOfSpeechTagging, enablePlaceRecognize, enableTranslatedNameRecognize, mergeNumberQuantifier, quickAtomSegment, seg, seg, seg2sentence, seg2sentence, segSentence, simpleAtomSegment

Methods inherited from class java.lang.Object

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

Constructor Detail

WordBasedSegment

public WordBasedSegment()

Method Detail

generateWord

protected static void generateWord(List<Vertex> linkedArray,
                                   WordNet wordNetOptimum)

对粗分结果执行一些规则上的合并拆分等等，同时合成新词网

Parameters:

linkedArray - 粗分结果

wordNetOptimum - 合并了所有粗分结果的词网

fixResultByRule

protected static void fixResultByRule(List<Vertex> linkedArray)

通过规则修正一些结果

Parameters:

linkedArray -

convert

protected static List<Term> convert(List<Vertex> vertexList)

将一条路径转为最终结果

Parameters:

vertexList -

Returns:

generateBiGraph

protected static Graph generateBiGraph(WordNet wordNet)

生成二元词图

Parameters:

wordNet -

Returns:

generateWordNet

protected void generateWordNet(WordNet wordNetStorage)

生成一元词网

Parameters:

wordNetStorage -

decorateResultForIndexMode

protected List<Term> decorateResultForIndexMode(List<Vertex> vertexList,
                                                WordNet wordNetAll)

为了索引模式修饰结果

Parameters:

vertexList -

wordNetAll -

speechTagging

protected static void speechTagging(List<Vertex> vertexList)

词性标注

Parameters:

vertexList -