This year out of four matchers registered for the complex track one matcher, AMLC, managed to match the consensus conference dataset. CANARD needs instances for matching which are missing in this dataset. XMap and KEPLER did not generate any complex correspondence.
The following table summarizes numbers of complex equivalence correspondences generated by AMLC and their types per each ontology pair. AMLC focused on two types of complex equivalence correspondences: those with attribute occurence restriction and those with attribute domain restriction.
| ontology pair | #AttributeOccurenceRestriction | #AttributeDomainRestriction | #complex equivalence correspondences |
|---|---|---|---|
| conference-ekaw | 2 | 9 | 11 |
| cmt-conference | 1 | 4 | 5 |
| cmt-ekaw | 4 | 3 | 7 |
The following table summarizes the evaluation resuts of AMLC on three ontology pairs.
| ontology pair | P | R | F-measure |
|---|---|---|---|
| conference-ekaw | 0.36 | 0.20 | 0.26 |
| cmt-conference | 0.4 | 0.22 | 0.28 |
| cmt-ekaw | 0.86 | 0.60 | 0.71 |
| average | 0.54 | 0.34 | 0.42 |
Considering that this is the first year of running the complex track and this dataset the achieved results are pretty promising. Particularly, AMLC managed to achieve high performance on the cmt-ekaw ontology pair. Congratulations!
The generated alignments by AMLC are available