Consensus building workshop

1
Consensus building workshop

• Conference track

2
Outline

• Introduction (ideas behind the track)
• Evaluation
• Discussion interesting mappings

3
Conference track - Features

• Broadly understandable domain Conference
  Organisation
• Free exploration by participants within 10
  ontologies
• No a priori reference alignment
• Participants 6 research groups

4
Conference track - Dataset
OWL, tool Protege
5
Conference track - Participants

• 6 participants
• Automs
• Coma
• OWL-CtxMatch
• Falcon
• HMatch
• RiMOM

6
Conference track - Goals

• Focus on interesting mappings and unclear
  mappings
• Why should they be mapped?
• Arguments against and for
• Which systems did discover them?
• Differences in similarity measures
• Underlying techniques?

7
Outline

• Introduction (ideas behind the track)
• Evaluation
• Discussion interesting mappings

8
Evaluation

• Processing all mappings by hand
• Assessment based on personal judgement of
  organisers (consistency problem)
• Tags TP, FP, interesting, ?, heterogenous
  mapping
• Types of errors and phenomena
• subsumption, inverse property, siblings, lexical
  confusion

9
Evaluation

• Subsumption mistaken for equivalence
• Author,Paper_Author
• Conference_Trip, Conference_part
• Inverse property
• has_author,authorOf
• Siblings mistaken for equivalence
• ProgramCommittee,Technical_commitee
• Lexical confusion error
• program,Program_chair
• Relation Class mapping
• has_abstract,Abstract
• Topic,coversTopic read_paper,Paper

10
Evaluation

• Some statistics as a side-effect of processing

11
Evaluation
12
Evaluation

• Distribution of similarity measures
• for True Positive Mappings and
• for False Positive Mappings

13
Evaluation
14
Evaluation
15
Evaluation
16
Evaluation
17
Record it!
18
Outline

• Introduction (ideas behind the track)
• Evaluation
• Discussion interesting mappings

19
Discussion

• Focus on interesting mappings and unclear
  mappings
• Why should they be mapped?
• Arguments against and for
• Which systems discover them?
• Differences in similarity measures
• Underlying techniques?

20
Mapping 1
Element1 Element1 Element2 Element2 Element2 Element2 Relation Relation
Person Person Confioushuman Confioushuman Confioushuman Confioushuman
Notes Notes semantically same semantically same semantically same semantically same semantically same semantically same
System System System System System System System System
Automs Coma Coma OWL-CtxMatch Falcon HMatch HMatch RiMOM
1.0 Iasted 1.0 Ekaw No No No No 0.7 confOf 0.63 Ekaw 0.81 Sigkdd 0.77 sofsem 0.7 confOf 0.63 Ekaw 0.81 Sigkdd 0.77 sofsem 1.0 PCS
21
Mapping 1
Iasted
confious
sigkdd
sofsem
ekaw
confOf
PCS
22
Mapping 2
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
OpenConfSurname OpenConfSurname OpenConfSurname confiouslast_name confiouslast_name confiouslast_name
Notes Notes Notes Both are datatype properties, the former with People as domain, the latter with human as domain Both are datatype properties, the former with People as domain, the latter with human as domain Both are datatype properties, the former with People as domain, the latter with human as domain Both are datatype properties, the former with People as domain, the latter with human as domain Both are datatype properties, the former with People as domain, the latter with human as domain
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
1.0 No No No No 1.0 1.0 No
23
Mapping 3
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
sofsemhas_the_last_name sofsemhas_the_last_name sofsemhas_the_last_name confiouslast_name confiouslast_name confiouslast_name
Notes Notes Notes Both are datatype properties, the former with Person as domain, the latter with human as domain Both are datatype properties, the former with Person as domain, the latter with human as domain Both are datatype properties, the former with Person as domain, the latter with human as domain Both are datatype properties, the former with Person as domain, the latter with human as domain Both are datatype properties, the former with Person as domain, the latter with human as domain
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
No 0.63 No No No 0.8 0.8 No
24
Mapping 4
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
ekaw PC_Member ekaw PC_Member ekaw PC_Member confOfMember_PC confOfMember_PC confOfMember_PC
Notes Notes Notes Change order of incompound names Change order of incompound names Change order of incompound names Change order of incompound names Change order of incompound names
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
1.0 No No No No No No 0.53
25
Mapping 4
confOf
ekaw
26
Mapping 5
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
ekawDocument ekawDocument ekawDocument confiousarticle confiousarticle confiousarticle
Notes Notes Notes Semantically same? Semantically same? Semantically same? Semantically same? Semantically same?
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
1.0 No No No No No No 1.0
27
Mapping 5
ekaw
confious
28
Mapping 8
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
cmtRejection cmtRejection cmtRejection OpenConfReject OpenConfReject OpenConfReject
Notes Notes Notes Both relates to process of assessment. But the former is a recommendation, the latter is a decision. So Both relates to process of assessment. But the former is a recommendation, the latter is a decision. So Both relates to process of assessment. But the former is a recommendation, the latter is a decision. So Both relates to process of assessment. But the former is a recommendation, the latter is a decision. So Both relates to process of assessment. But the former is a recommendation, the latter is a decision. So
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
No 0.29 No No 0.94 No No No
29
Mapping 8
OpenConf
cmt
30
Mapping 11
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
ekawLocation ekawLocation ekawLocation Place Place Place
Notes Notes Notes Semantically same? Both are at the highest level of hierarchy. But Location maybe more general than Place what about City? Semantically same? Both are at the highest level of hierarchy. But Location maybe more general than Place what about City? Semantically same? Both are at the highest level of hierarchy. But Location maybe more general than Place what about City? Semantically same? Both are at the highest level of hierarchy. But Location maybe more general than Place what about City? Semantically same? Both are at the highest level of hierarchy. But Location maybe more general than Place what about City?
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
No No No No No iasted 0.8 sigkdd 0.8 iasted 0.8 sigkdd 0.8 No
31
Mapping 11
ekaw
iasted
sigkdd
Location is domain of properties
locationOf Location is range of properties
heldIn
iastedPlace is domain of properties
is_equipped_by sigkddPlace is range of
properties can_stay_in
Asserted conditions for iastedPlace
32
Mapping 12
Element1 Element1 Element1 Element2 Element2 Element2 Relation Relation
sofsemreviews sofsemreviews sofsemreviews ekawhasReview ekawhasReview ekawhasReview
Notes Notes Notes DomainOf(hasReview)Paper,rangeOf(hasReview)Review DomainOf(reviews)Review,rangeOf(reviews)Reviewed_contribution Inverse property phenomena, useful? DomainOf(hasReview)Paper,rangeOf(hasReview)Review DomainOf(reviews)Review,rangeOf(reviews)Reviewed_contribution Inverse property phenomena, useful? DomainOf(hasReview)Paper,rangeOf(hasReview)Review DomainOf(reviews)Review,rangeOf(reviews)Reviewed_contribution Inverse property phenomena, useful? DomainOf(hasReview)Paper,rangeOf(hasReview)Review DomainOf(reviews)Review,rangeOf(reviews)Reviewed_contribution Inverse property phenomena, useful? DomainOf(hasReview)Paper,rangeOf(hasReview)Review DomainOf(reviews)Review,rangeOf(reviews)Reviewed_contribution Inverse property phenomena, useful?
System System System System System System System System
Automs Coma OWL-CtxMatch OWL-CtxMatch Falcon HMatch HMatch RiMOM
No No 1.0 1.0 No No No No
33

• Call for contribution to our dataset

34
Thank you for your participation!