org.apdplat.word.analysis

Class JaccardTextSimilarity

java.lang.Object

org.apdplat.word.analysis.TextSimilarity

org.apdplat.word.analysis.JaccardTextSimilarity

All Implemented Interfaces:

Similarity, SimilarityRanker

public class JaccardTextSimilarity
extends TextSimilarity

文本相似度计算 判定方式：Jaccard相似性系数（Jaccard similarity coefficient） ，通过计算两个集合交集的大小除以并集的大小来评估他们的相似度 算法步骤描述： 1、分词 2、求交集（去重），计算交集的不重复词的个数 intersectionSize 3、求并集（去重），计算并集的不重复词的个数 unionSize 4、2中的值除以3中的值 intersectionSize/(double)unionSize 完整计算公式： double score = intersectionSize/(double)unionSize;

Author:

杨尚川

Field Summary

Fields inherited from class org.apdplat.word.analysis.TextSimilarity

filterStopWord, LOGGER

Fields inherited from interface org.apdplat.word.analysis.Similarity

thresholdRate

Constructor Summary

Constructors

Constructor and Description
JaccardTextSimilarity()

Method Summary

Modifier and Type	Method and Description
static void	main(String[] args)
protected double	scoreImpl(List<Word> words1, List<Word> words2) 判定相似度的方式：Jaccard相似性系数

Methods inherited from class org.apdplat.word.analysis.TextSimilarity

setSegmentationAlgorithm, similarScore, similarScore, taggingWeightWithWordFrequency, toFastSearchMap

Methods inherited from class java.lang.Object

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

Methods inherited from interface org.apdplat.word.analysis.SimilarityRanker

rank, rank

Methods inherited from interface org.apdplat.word.analysis.Similarity

isSimilar, isSimilar, isSimilar, isSimilar, similarScore, similarScore

Constructor Detail

JaccardTextSimilarity

public JaccardTextSimilarity()

Method Detail

scoreImpl

protected double scoreImpl(List<Word> words1,
                           List<Word> words2)

判定相似度的方式：Jaccard相似性系数

Specified by:

scoreImpl in class TextSimilarity

Parameters:

words1 - 词列表1

words2 - 词列表2

Returns:

相似度分值

main

public static void main(String[] args)