YAM++ - A combination of graph matching and machine learning - - PowerPoint PPT Presentation
YAM++ - A combination of graph matching and machine learning approach to ontology alignment task DuyHoa Ngo, Zohra Bellahsene Amir Naseri Knowledge Engineering Group 28. Januar 2013 Introduction An Ontology is a formal specification
2
3
4
5
6
7
8
9
10
11
12
13
14
15
1, O 2: ontologies
0 = {(e 1, e 2, ≡, w 0)}: initial mappings
1, e 2, ≡, w 1)}: result mappings
1 ← Transform (O 1)
2 ← Transform (O 2)
1, G 2)
0)
s)
16
s is used to produce the final result
17
18
19
20
21
element and M structure are set of mappings found by element and structure
22
element = {(e i, e j, ≡, 1)}
structure = {(e p, e q, ≡, c s) , c s ∈ (θ s, 1]}
final = {(e 1, e 2, ≡, c) , c ∈ [0, 1]}
s) : m ∈ M structure ∩ M element
element , θ, M structure ,(1 – θ))
final
final )
final
23
24
element is smaller than the
structure
element and M structure
25
26
1), (x 1, y)} ∈ A and x 1 ∈ Desc (x), y 1 ∈ Desc (y) →
1), (x 1, y) are inconsistent and will be removed
27
1, p 2) ∈ A and { Doms (p 1) x Doms (p 2) ∩ A = Ø } and
1) x Rans (p 2) ∩ A = Ø } →
1, p 2) is inconsistent and will be removed
28
29
30
i|) / (Σ |A i|),
i|) / (Σ |R i|),
m) = (2 * H p * H r) / (H p + H r) .
i|: number of correct mappings
i|: total number of mappings
i|: number of reference mappings
31
32
33
34
35
measure in 3 ways
0.5: recall more important than precision
1: recall and precision equally important
2: precision more important than recall
36
37
38