“Evaluating some Heurystics to Find Hyponyms between Ontologies”
This page contains the results of the experiments described in the chapter book “Evaluating some Heurystics to Find Hyponyms between Ontologies” by Ignacio Huitzil, Fernando Bobillo, Eduardo Mena, Carlos Bobed and Jesús Bermúdez.
In this link, we can see an extended version of the detailed results of the work [1]
- - Experimental results (Translate and WordNet).
Dataset:
The datasets can be directly obtained from OAEI 2009 file_download, OAEI 2011 file_download, ORE 2015 file_download, OAEI 2016 file_download.
Statistics of the datasets:
The next Table shows some statistical data of each dataset considered in our experiments: Total number of examples in the dataset (TOT), examples correctly processed within a timeout (OK), average number of Classes (C), average number of subClasses (sub C), average number of pairs of Classes (pairs C), number of ontologies with Object Properties (OOP), average number of Object Properties (OP), number of ontologies with Data Properties (ODP), average number of Data Properties (DP), average number of subObjectProperties (sub OP), average number of subDataProperties(sub DP), average number of pairs of Object Properties (pairs OP), and average number of pairs of Data Properties (pairs DP).
| Dataset | TOT | OK | C | sub C | pairs C | OOP | OP | ODP | DP | sub OP | pairs OP | sub DP | pairs DP |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| OAEI 2009 | 30 | 23 | 77 | 48 | 1424 | 23 | 52 | 23 | 89 | 0 | 0 | 0 | 0 |
| OAEI 2009* | 3 | 3 | 72 | 33 | 1259 | 3 | 51 | 3 | 92 | 0 | 0 | 0 | 0 |
| OAEI 2011 | 12 | 9 | 154 | 84 | 6006 | 3 | 17 | 3 | 14 | 0 | 0 | 0 | 0 |
| OAEI 2011* | 6 | 3 | 56 | 18 | 597 | 3 | 52 | 3 | 43 | 0 | 0 | 0 | 0 |
| OAEI 2016 | 3 | 3 | 101 | 11 | 2619 | 3 | 73 | 3 | 51 | 0 | 0 | 0 | 0 |
| ORE 2015 | 1920 | 1013 | 792 | 857 | 1834180 | 925 | 44 | 375 | 9 | 25 | 6600 | 3 | 1373 |
| ORE 2015* | 6 | 6 | 78 | 88 | 6772 | 6 | 57 | 4 | 19 | 28 | 4237 | 0.5 | 604 |
Experimental Results, translate (Transl.) and Wordnet (Word.):
- Datasets Status The next Table shows the datasets status with respect to language for on experiments:
| Dataset | Total | No English | Alphanumeric or Unreadable |
|---|---|---|---|
| OAEI 2009 | 30 | 3 | 7 |
| OAEI 2009* | 0 | 3 | 0 |
| OAEI 2011 | 12 | 0 | 0 |
| OAEI 2011* | 3 | 0 | 0 |
| ORE 2015 | 1013 | 6 | 491 |
| ORE 2015* | 0 | 6 | 0 |
| OAEI 2016 | 3 | 0 | 0 |
-Metrics for shared roles on each dataset The next Table shows the result of the measures related to shared roles, from a) to d). For each dataset, we show the total number of examples found (Sum), the total number of ontologies with at least one example (#Onts), and the average percentage of examples (Mean%). Translation dictionaries (T), Wortnet (W) and both (TW).
| Criteria | b Vs. c | b Vs. d | |||||||
|---|---|---|---|---|---|---|---|---|---|
| Dataset | Item | a | b | c | d | #b> | Dif | #b> | Dif |
| OAEI 2009 (C) W | Sum | 354 | 460 | 640 | 1989 | ||||
| #Onts | 10 | 22 | 22 | 10 | 3 | -14 | 12 | 2 | |
| Mean % | 32 | 42 | 58 | 6 | |||||
| OAEI 2009* (C) T | Sum | 0 | 50 | 50 | 0 | ||||
| #Onts | 0 | 2 | 2 | 0 | 0 | 0 | 2 | 2 | |
| Mean % | 0 | 50 | 50 | 0 | |||||
| OAEI 2009* (C) TW | Sum | 0 | 42 | 58 | 0 | ||||
| #Onts | 0 | 2 | 2 | 0 | 0 | -2 | 2 | 2 | |
| Mean % | 0 | 42 | 58 | 0 | |||||
| OAEI 2011 (C) W | Sum | 0 | 2 | 762 | 0 | ||||
| #Onts | 0 | 2 | 9 | 0 | 0 | -8 | 2 | 2 | |
| Mean % | 0 | 0.3 | 100 | 0 | |||||
| OAEI 2011* (C) W | Sum | 0 | 2 | 51 | 0 | ||||
| #Onts | 0 | 2 | 3 | 0 | 0 | -3 | 2 | 2 | |
| Mean % | 0 | 4 | 100 | 0 | |||||
| OAEI 2016 (C) W | Sum | 0 | 12 | 21 | 1 | ||||
| #Onts | 0 | 3 | 3 | 1 | 1 | -1 | 3 | 3 | |
| Mean % | 0 | 36 | 64 | 0.01 | |||||
-Metrics for string affixes on each dataset The next Table shows the result of our measures related to entity names, cases from a e) to g).
| Criteria | e Vs. g | |||||
|---|---|---|---|---|---|---|
| Dataset | Item | e | f | g | #e> | Dif |
| OAEI 2009 (C) W | Sum | 51 | 20 | 452 | ||
| #Onts | 15 | 12 | 23 | 0 | -23 | |
| Mean % | 5 | 2 | 1.4 | |||
| OAEI 2009* (C) T | Sum | 4 | 0 | 48 | ||
| #Onts | 2 | 0 | 3 | 0 | -3 | |
| Mean % | 4 | 0 | 1.3 | |||
| OAEI 2009* (C) TW | Sum | 6 | 2 | 211 | ||
| #Onts | 2 | 0 | 3 | 0 | -3 | |
| Mean % | 6 | 2 | 5.7 | |||
| OAEI 2011 (C) W | Sum | 77 | 0 | 242 | ||
| #Onts | 8 | 0 | 9 | 1 | -6 | |
| Mean % | 10 | 0 | 0.5 | |||
| OAEI 2011* (C) W | Sum | 9 | 0 | 27 | ||
| #Onts | 2 | 0 | 3 | 0 | -3 | |
| Mean % | 17 | 0 | 1.5 | |||
| OAEI 2016 (C) W | Sum | 17 | 1 | 146 | ||
| #Onts | 3 | 1 | 3 | 0 | -3 | |
| Mean % | 52 | 3 | 2 | |||
| ORE 2015 (C) W | Sum | 37802 | 794 | 816489 | ||
| #Onts | 495 | 102 | 680 | 69 | -538 | |
| Mean % | 4.3 | 0.1 | 0.04 | |||
| ORE 2015 (OP) W | Sum | 50 | 141 | 11860 | ||
| #Onts | 49 | 46 | 518 | 0 | -518 | |
| Mean % | 0.2 | 0.6 | 0.2 | |||
| ORE 2015 (DP) W | Sum | 6 | 11 | 2508 | ||
| #Onts | 6 | 3 | 204 | 0 | -204 | |
| Mean % | 0.20 | 0.37 | 0.2 | |||
| ORE 2015* (C) T | Sum | 39 | 0 | 49 | ||
| #Onts | 5 | 0 | 6 | 1 | -4 | |
| Mean % | 7.4 | 0 | 0.12 | |||
| ORE 2015* (C) TW | Sum | 39 | 1 | 70 | ||
| #Onts | 5 | 1 | 6 | 1 | -4 | |
| Mean % | 7.4 | 0.2 | 0.2 | |||
| ORE 2015* (OP) T | Sum | 0 | 0 | 40 | ||
| #Onts | 0 | 0 | 6 | 0 | -6 | |
| Mean % | 0 | 0 | 0.16 | |||
| ORE 2015* (OP) TW | Sum | 0 | 0 | 183 | ||
| #Onts | 0 | 0 | 6 | 0 | -6 | |
| Mean % | 0 | 0 | 0.8 | |||
| ORE 2015* (DP) T | Sum | 1 | 0 | 22 | ||
| #Onts | 1 | 0 | 3 | 0 | -3 | |
| Mean % | 33 | 0 | 0.6 | |||
| ORE 2015* (DP) TW | Sum | 1 | 0 | 129 | ||
| #Onts | 1 | 0 | 4 | 0 | -4 | |
| Mean % | 33 | 0 | 3.6 | |||
The detailed results about each dataset (Translation, WordNet) can be seen in the next link.
[1] Huitzil, I., Bobillo, F., Mena, E., Bobed, C., Bermúdez, J. Some reflections on the discovery of hyponyms between ontologies. Proceedings of the 21st International Conference on Enterprise Information Systems (ICEIS 2019). vol. 2, pp. 130-140. SCITEPRESS (2019), Result website