Paper: Fuzzy Semantic Distance MeasureS Between Ontology Concepts
Profile
From: IEEE 2004
Author: Valerie Cross
E-mail: crossv@mouhio.edu
Idea Abstract
NOTES:
: Similarity Between Concepts
: Distance between concepts
: Relation Between Concepts
Way One
EXPRESSION
(Expression 1)
,
: Different Concepts in One ONTOLOGY
: The maximum defth of the amount
Sketch Map
Di
C1
C2
Chart 1
Way Two
EXPRESSION
(Expression 2)
: The Length from the root to
: The Length from the root to
: The Length from the root to
, Which is the nearest Parent Node for
and
Sketch Map
C1
C2
C3
N3
N1
N2
Chart 2
Way Three
EXPRESSION
(Expression 3)
Sketch Map
(Just As the Way Two.)
Way Four
Thinking
Two concepts are semantically related if they are connected by a path that is not longer than an arbitrary fixed constant C and whose direction does not change too often (where d represents the number of changes in path direction and k is another constant.)
EXPRESSION
(Expression 4)
: An Arbitrary Fixed Constant
,
: Two DiffERENT Concepts
D: The Number of Changes in Path Direction
K: a constant
Sketch Map
C1
C2
C3
Chart 3
Way Five
Thinking
Employ The Different Kinds of Linking Relationship WITHIN WordNet. Each Edge Maps To Two Inverse Relations. Each Type of Relations R Has A Weight Range Between ITS OWN
and
. The actual value in That Range for
Depends on
, The Number of Relations of Type
leaving node X. This value referred to as type specific fanout (TSF) factor incorporates the dilution of the strength of connotation between a source and target node as a function of the number of like relations that the source node has. This factor reflects that asymmetry Might Exist Between The Two Nodes So That The Stregth of Connotation In One Direction Differs. Transfiguration: Different connection relationships are used in WordNet. Each side corresponds to the two opposite relationships, each relationship R corresponds to a range of weights existing, minimum. R's accuracy value
Decided, the number of X node R-class relationships is the number. This and type detail factors are integrated with the value of each connotation, and these connotations exist in resource and target nodes as a numeric function, and these values are similar to the various relationships of the resource node. This factor exhibits asymmetry that may exist in two nodes, so that the length of the connotation in the direction is different.
EXPRESSION
(Expression 5)
(Expression 6)
Way Six
Thinking
The information-theoretic models of semantic similarity add to the information already present in the network by using a qualitatively different, knowledge source. The groundwork for much of this research is founded on the insight that conceptual similarity between two concepts C1 and C2 may be judged ............ .י.
Translation: The theoretical model of the semantic similarity is added to the information that has been expressed by the network using a different quality knowledge resource. Based on these studies have shown that similarity between the two concepts can be determined by their share of information, the more information they share, the greater the degree of similarity. Various methods are used to measure the content of the information.
EXPRESSION
(Expression 7)
: In A Specific Corpus, The Probability of Encountering An Instance of C3, Which Is The Information Content of The Lowest Superconcept and Subsumes C1, C
2 in
The hierarchy.c3 is also regarded as the MOST SPECIFIC COMMON ANCESTOR OF C1 and C2.
(Expression 8)
: The Probability That A Randomly SELECTED Object X1 10.
C3: The Most Specific Concept That Subsumes Both C1 and C2. (Expression 9)
Note:. Study has been made in the roles of the density and depth components Then they concluded that they are not major factors in the overall edge weight and shifted their focus to the link-strength factor And this measure is but a different arithmetic combination. Of The Same Terms in Expression 8.
IDEA Conclusion
Between Two Concepts
THESE MeasureS Have Been Classified Into Two PRIMARY Approaches:
l The First Uses Distance with Respect to Path Length Between The Two Concepts Wtem (The Main Link Has Been The IS-a TaxOmic Link).
.
THESE TWO MEASURES May Be Viewed As:
(Expression 10)
: The set of is-a links from the root to concept c1
: The set of is-a links from the root to concept c2
For Example, IF
=
= 1/2, THEN S BECOMES:
(Expression 11)
C1
C2
C1
Chart 4
Between TOE Concept Sets
(Expression 12)
,
: Two Different Concept Sets
One Example, Which Changes from Expression 12:
(Expression 13)
ME
