Leaderboard
Here are the standings of the final submissions:Hybrid KGQA Results
| Team | Exact Match | F-Score |
|---|---|---|
| fokam_emeric (Fomubad Borista Fondi, Azanzi Jiomekong Fidel) | 0.4359 | 0.4505 |
| nsteinmetz (Kanchan Shivashankar and Nadine Steinmetz) | 0.3205 | 0.4070 |
Honorable mention
Submitted on an different dataset split
| Team | Exact Match | METEOR |
|---|---|---|
| Sefika (Sefika Efeoglu, Nikolas Rauscher, Erik Rubinov, Yuxin Xue and Sonja Schimmler) | 0.4890 | 0.5171 |
UPDATE
UPDATE
Task Description
In the previous version of the Scholarly QALD Challenge at ISWC 2023, 7 teams took part in the tasks of entity linking and question answering over the DBLP and ORKG Knowledge Graphs (KG). We releasd two datasets, each catering to one KG, where participants were required to fetch the answer for a given question from the corresponding KG.
This year, we introduce a more challenging task of Hybrid-QA, which will require participants to query multiple sources of knowledge to fetch the right answer for a given question. The domain of the challenge remains scholarly data, however unlike last year, participants will have to find the answers from multiple textual sources (e.g., Wikipedia) and Knowledge Graphs (e.g., DBLP).
By taking part in the challenges sub-tasks, participants can explore and will experience a versatile range of academic and industrial approaches and applica- tions through the multi-faceted workshop format. We will try to bridge the gap between academia and industry to attract junior as well as senior researchers from both worlds leading to a memorable experience. We target to publish system descriptions as proceedings published by it - Information Technology Journal.
In our second iteration at ISWC 2024, we have the following task:
Task: Hybrid-QA — Hybrid Question Answering over multiple Knowledge Graphs and Text Sources: For a given question, participants have to consider multiple scholarly sources for finding the correct answer, e.g., DBLP KG, Wikipedia text, OpenaAlex KG etc.
