Call For Papers/How to Apply

Challenge Goal

In a survey conducted between 2010 and 2012, 77.1% of accidents involving people aged 65 years and above occurred in their homes. This finding highlights the importance of ensuring the safety of older adults in their homes. We can attract more attention by focusing on this issue in concrete and familiar settings. The "Knowledge Graph Reasoning Challenge for Social Issues" provides a standardized dataset of knowledge graphs and videos that can be used to develop safety technologies for older adults. These technologies can include cyber-physical systems, AI speakers, and AR systems, which can help occupants, home designers, and safety engineering professionals detect accident risks and provide safer alternatives. This year's task is to extract statistical information on interactions with actions and objects that lead to risks in daily life, which will help achieve these goals.

In the workshop, we would like to organize the Second International Knowledge Graph Reasoning Challenge (IKGRC 2024) to discuss a wider variety of reasoning technologies using multimodal data.

Challenge Task

The provided dataset contains video data that shows different daily life actions in VirtualHome. These videos are transformed into knowledge graphs that provide information about the person performing the action, what is involved in the action, and the resulting state or position of the object.
The task is to extract statistical information on interactions with actions and objects that lead to risks in daily life. We have prepared 70 episodes and eight specific questions. The task is to answer eight questions for each of the 70 scenarios. All data is available on our github repository ( https://github.com/KnowledgeGraphJapan/KGRC-RDF/tree/kgrc4si).
We are soliciting entries for the dataset in the following two divisions.

• Phase1 (for beginners): The knowledge graph provided contains all the information and can be searched with SPARQL to answer each question.
• Phase2 (for advanced): The knowledge graph provided is partially anonymized information, and each question must be answered by predicting that information from the knowledge graph or video.

For more information on the task, please see https://ikgrc.org/2024/application.html

Submissions

• 1. Please submit(*) a short paper (min. 2 pages, max 8 pages, IEEE format) with an outline of each of the submitted categories by [ December 25, 2023]. Systems and tools may be under development. After peer review, accepted papers will be published in the Proceedings of the ICSC.
  (*) The EaasyChair URL for the IKGRC 2024 is https://easychair.org/conferences/?conf=ikgrc2024.
• 2. If your paper is accepted, you will be required to submit the camera ready by 30th December
• 2. If your paper is accepted, you also will be required to send us* materials explaining your system, tools, and ideas by January 22th, 2024 for the final document review. *Please send the material to kgrc@knowledge-graph.jp via email.
• 3. On the day of the challenge, you will be asked to present your system, tool, or idea with a demonstration (online participation is possible). Prizes will be awarded to the best entries.
  *Time for the presentation is 20 min. including 5 min. QA.

Evaluation Criteria (Details to be announced)

• Common for all tasks: Accuracy rate, technical ingenuity, method of inference, expanded knowledge.
• Evaluation based on the paper submitted in advance and the presentation on the day.
• It is planned to conduct a peer review on the day of the event. At least one member from each team should participate in the workshop (to be held on February 5-7, 2024).

Contact