For task 1, all our evaluations ran on a server with an AMD Ryzen Threadripper 2990WX 32-Core Processor and 128GB of RAM. The server ran on a 5.3.0-46-generic #38 18.04.1-Ubuntu SMP operating system. We use Java 1.8 and OWL API 5.1.11 in the experiments. We set the heap space to 24GB. We report the average time taken in 2 independent runs for the classification task.
Task-1
We got 2 submissions in this category (arranged in the order of ranking). We evaluated each ontology in terms of the time-taken(in seconds) by different OWL 2 DL reasoners (HermiT 1.4.5.456, JFact 5.0.05, Konclude 0.7.0, Openllet 2.6.46 , and Pellet 2.3.6).LE (Loading Error), PE (Parsing Error), ME (Memory Error), TO (Time-out)
| Submissions | Submitted Ontologies | Expressivity | Total Axioms Count | Logical Axioms Count | Openllet | Konclude |
| FIBO | FIBO (Prod) | SROIQ(D) | 90186 | 50121 | 10800 | ME |
| FIBO (Dev) | SROIQ(D) | 98810 | 53591 | 28800 | ME | |
| BAALL | FOD_Dish_all | SHIF(D) | 35419 | 13348 | TO | 0.050 |
| FOD_Small_all | SHF(D) | 20189 | 6610 | Null Pointer Exception | 0.030 | |
| Ingredient_down | SROIQ(D) | 537 | 243 | 195.20 | 0.002 | |
| Ingredient_i032to001 | SROIQ(D) | 373 | 202 | 130.23 | 0.002 | |
| FOD_Product_all | SHIF(D) | 29270 | 10889 | Null Pointer Exception | 0.048 | |
| FOD_Small_i032to001 | SROIQ(D) | 20196 | 6631 | Null Pointer Exception | 0.002 | |
| Ingredient_i016to001 | SROIQ(D) | 373 | 202 | 0.13 | 0.002 | |
| Ingredient_i064to001 | SROIQ(D) | 537 | 243 | 0.14 | 0.002 |
Task-2
Results of the submitted Neuro-symbolic reasoners. These results are the same as reported by the authors in the paper.Memory Networks for RDFS reasoning: Experiments
Improving Transitive Embeddings in Neural Reasoning Tasks via Knowledge-Based Policy Networks
