edu.hust.go.model
Class Similarity

java.lang.Object
  |
  +--edu.hust.go.model.Similarity

public class Similarity
extends java.lang.Object


Field Summary
static byte CAO
           
static byte LIN
           
static byte oldZZL
           
static byte RESNIK
           
static byte ZZL
           
 
Constructor Summary
Similarity()
           
 
Method Summary
static float evaluate(edu.hust.go.model.GoGraphModel graphModel, java.lang.Object obj1, java.lang.Object obj2, byte type)
          evaluate the similarity of two nodes in directed acylic graph
 
Methods inherited from class java.lang.Object
clone, equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait
 

Field Detail

CAO

public static final byte CAO
See Also:
Constant Field Values

LIN

public static final byte LIN
See Also:
Constant Field Values

RESNIK

public static final byte RESNIK
See Also:
Constant Field Values

ZZL

public static final byte ZZL
See Also:
Constant Field Values

oldZZL

public static final byte oldZZL
See Also:
Constant Field Values
Constructor Detail

Similarity

public Similarity()
Method Detail

evaluate

public static float evaluate(edu.hust.go.model.GoGraphModel graphModel,
                             java.lang.Object obj1,
                             java.lang.Object obj2,
                             byte type)
evaluate the similarity of two nodes in directed acylic graph

Parameters:
graphModel - the graph that contains the two nodes
obj1 - one of the two nodes
obj2 - one of the two nodes
type - Similarity.CAO: basen on information content Similarity.LIN: basen on information content, see Lin D. An information-theoretic definition of similarity. In: Proceedings of 15th International Conference on Machine Learning, San Francisco US, 1998, 296-304 Similarity.RESNIK: based on information content, see Resnik P. Using information content to evaluate semantic similarity in a taxonomy. In: Proceedings of the 14th International Joint Conference on Artificial Intelligence, Montreal Canada, 1995, 448-453 Similarity.ZZL: refined ZZL based on distance, the LCA(latest common ancestor) is one with maximal information content in the of the same distance LCA Similarity.oldZZL: based on distance, see Zhong JW, Zhu HP, Li JM, Yu Y. Conceptual Graph Matching for Semantic Search. In: Proceedings of the 10th International Conference on Conceptual Structures: Integration and Interfaces, 2002, 92-196
Returns:
the semanic similarity value