Daily Arxiv

This is a page that curates AI-related papers published worldwide.
All content here is summarized using Google Gemini and operated on a non-profit basis.
Copyright for each paper belongs to the authors and their institutions; please make sure to credit the source when sharing.

Integrating Activity Predictions in Knowledge Graphs

Created by
  • Haebom

Author

Forrest Hare Alec Sculley, Cameron Stockton

Outline

This paper argues that ontologically structured knowledge graphs can play a crucial role in predicting future events, leveraging the Basic Formal Ontology (BFO) and the Common Core Ontology (CCO). We present a method for organizing and retrieving data, such as the movement paths of fishing vessels, into a knowledge graph to generate a Markov chain model, which can then be used to predict future states based on the vessel's past paths. To complete the necessary structural semantics, we introduce the term "spatiotemporal instant," critique existing probabilistic ontological models of the future, and propose an alternative perspective that considers at least some probabilities to be related to actual process profiles to better capture the dynamics of real-world phenomena. Finally, we demonstrate how Markov chain-based probability calculations can be integrated into the knowledge graph to support further analysis and decision-making.

Takeaways, Limitations

•
Takeaways:
◦
Presenting the possibility of future prediction using a knowledge graph with an ontology structure.
◦
Presenting a systematic data organization and retrieval method based on BFO and CCO.
◦
Predicting future states using Markov chain models
◦
Improving semantic completeness through the introduction of the concept of 'spatiotemporal moment'.
◦
A new ontological perspective on probability
◦
Integrated analysis and decision support based on knowledge graphs
•
Limitations:
◦
Further research is needed on the real-world applicability and scalability of the proposed methodology.
◦
Need to verify generalizability to various types of data and complex systems
◦
The need for philosophical and empirical examination of the proposed new ontological perspective on probability.
◦
Consideration should be given to the limitations of Markov chain models (e.g., poor accuracy in long-term predictions).
👍
Made with Slashpage