The excitement around AI in 2023 was significantly driven by ChatGPT, which showcased recent advancements in AI. Although many were still wrestling with its functionality and potential applications, it is clear that AI has entered the mainstream. However, AI and related technologies still have a considerable journey ahead before they align with science-fiction visions of technological singularity; when all cyber security firewalls fail, the world will be populated with humans augmented with machines [Gawdat, 2023], or robots run out of control [Osawa et al., 2022].
In Computer Science, “an ontology defines a set of representational primitives with which to model a domain of knowledge or discourse” [Gruber,2009]. These are usually concepts, categories, and relationships within the Data-Information-Knowledge-Wisdom (DIKW) space. In other words, ontologies provide a formalised way to understand and organise content within this framework.
In this special issue of the AIA Journal, we expand the work of the MOVE community [R. Polovina et al., 2022], by exploring how ontologies are advancing AI and enhancing its applications, emphasising their significance in shaping the future of AI. We invite papers examining ontologies and similar methods for advancing AI in theory and practice.
AI applications are pervasive—they impact all types of organisations: enterprises, governments, non-governmental organisations (NGOs), societies, and human endeavours in general. It has been recognised that ontologies (and ontology-like methods) may benefit organisations at multiple levels [Baldazzi et al., 2023; Earley, 2020]. Thus, we also welcome papers that explore how contemporary organisations can use ontologies to better adapt to manage these disruptive technologies and execute their strategies through Enterprise or Endeavour Architectures or other relevant fields [R. Polovina & S. Polovina, 2022]. It is also crucial to explore ways to prevent AI from inhibiting societal development and to ensure that these technologies contribute positively to societal well-being [Viunesa et al., 2022].
Ontologies have been linked to AI in several significant ways, including, but not limited to:
Knowledge Representation and Sharing: By supporting formal DIKW representation, ontologies enable AI to understand, interpret, reason, and manipulate DIKW content. For example, Formal Concept Analysis (FCA) is a mathematical framework that derives concept hierarchies or formal ontologies from data. FCA can be pivotal in fields like Machine Learning (ML) and knowledge representation and reasoning. Therefore, FCA plays a significant role in the AI ecosystem. In general, the formal representations provided by ontologies facilitate content sharing within the DIKW space on multiple levels: among digital systems, digital systems and humans, or societies [Baxter et al., 2022; R. Polovina R., 2022; S. Polovina, 2013].
Semantic Modeling: By defining the semantics of concepts and their relationships, ontologies enable a deeper understanding of the content in the DIKW space. This is important in natural language processing and information retrieval because AI must interpret the content, context, and meaning more accurately. Semantic modelling includes, for example, Knowledge Graphs and Business Data Graphs (e.g., SAP Graph). These graphs are essential to the AI ecosystem and, hence, AI applications. They integrate metadata from disparate sources, enabling more effective and intelligent use of the hitherto hidden semantics in data [Baldazzi et al., 2023; Liu et al., 2023; SAP, n.d.].
Domain Expertise: Ontologies capture domain-specific DIKW, enabling AI to operate in specialised, vertical areas such as healthcare [Filice and Khan, 2021], finance [Bunell et al., 2021], or manufacturing [Naqvi et al., 2024]. This domain expertise enhances AI's ability to provide relevant and accurate DIKW within these vertical sectors.
Interoperability: Ontologies facilitate the integration of DIKW from diverse sources by offering a common vocabulary and framework. This unifies the DIKW landscape and enables AI to combine and analyse DIKW content [Swar et al., 2022]. The results are semantic clarity, reduced integration costs, and improved governance.
Complex Queries: Ontologies support the refinement of complex queries, enabling AI to handle these queries with greater accuracy [Kamran and Sheraz, 2018].
Automated Reasoning: Ontologies support automated reasoning, allowing the inference of new knowledge from existing DIKW content. This capability enables AI decision-making, problem-solving, and generating insights [Schneider and Šimkus, 2020].
Enhanced Communication, Cooperation and Collaboration: Ontologies may overcome heterogeneous taxonomies. In other words, ontologies facilitate a common understanding of terms and concepts, facilitating communication between heterogeneous AI systems, AI and humans, and among humans themselves [Earley, 2020]. This grounding empowers interoperability, cooperation, and collaboration [R. Polovina & S. Polovina, 2022].
Explication of knowledge: Ontologies may facilitate drawing intelligible explanations in neuro-symbolic AI by establishing correspondences between neural models and logical representation [Confalonieri & Guizzardi, 2024]. This feature is fundamental to reference modelling, common-sense reasoning, knowledge refinement and complexity management. In general, ontologies enable content sharing within the DIKW space.
The above advances AI by underpinning the formal and informal conceptual structures that AI tools can use to better reason, move forward their state-of-the-art, and bring AI applications to life.
We invite you, our readers and members of the MOVE community, to contribute to this ongoing dialogue by submitting your research, insights, and innovative applications of ontologies in AI. While we welcome all submissions, we are giving priority to those submitted through the AIA Journal. Your contributions are crucial in advancing our collective understanding and pushing the boundaries of AI. Join us in driving forward the transformative potential of ontologies in artificial intelligence by sharing your work today.
Please visit our Call for Papers for special issue of the AI and Applications page: https://ojs.bonviewpress.com/index.php/AIA/SI_MOVE
You could also submit your other contributions (e.g., abstracts, short papers) to the MOVE community by following this link: https://login.easychair.org/cfp/MOVE2024
We are looking forward to your original papers, perspectives and contributions.
Sincerely,
MOVE Organization Committee
References
Baldazzi, T., Bellomarini, L., Ceri, S., Colombo, A., Gentili, A., & Sallinger, E. (2023). Fine-tuning large enterprise language models via ontological reasoning. In Lecture Notes in Computer Science (pp. 86–94). https://doi.org/10.1007/978-3-031-45072-3_6
Baxter, M., Polovina, S., Kemp, N., & Laurier, W. (2022). Underpinning layered Enterprise Architecture development with Formal Concept Analysis. In Communications in Computer and Information Science (pp. 101–113). https://doi.org/10.1007/978-3-031-22228-3_6
Bunnell, L., Osei-Bryson, K., & Yoon, V. Y. (2021). Development of a consumer financial goals ontology for use with FinTech applications for improving financial capability. Expert Systems With Applications, 165, 113843. https://doi.org/10.1016/j.eswa.2020.113843
Confalonieri, R., Guizzardi, G. (2024). On the multiple roles of Ontologies in explanations for Neuro-symbolic AI. In A. Mileo(Ed.), Neurosymbolic AI. Available at https://www.neurosymbolic-ai-journal.com/content/issues
Earley, S. (2020). AI-Powered Enterprise–Harness the Power of Ontologies to Make Your Business Smarter, Faster and More Profitable. Life Tree Media Ltd. ISBN 978-1-928055-51-8 (PDF)
Filice, R.W., Kahn, C.E. Jr. (2021). Biomedical Ontologies to guide AI development in Radiology. J Digit Imaging. 2021 Dec; 34(6):1331-1341. doi: 10.1007/s10278-021-00527-1. Epub 2021 Nov 1. Erratum in: J Digit Imaging. 35(5):1419. doi: 10.1007/s10278-021-00568-6. PMID: 34724143; PMCID: PMC8669056.
Gawdat, M. (2023). Scary Smart: The Future of Artificial Intelligence and How You Can Save Our World. Bluebird Books for Life.
Groumpos, P. P. (2023). A critical historic overview of artificial intelligence: issues, challenges, opportunities and threats. Artificial Intelligence and Applications, 1(4), 197–213. https://doi.org/10.47852/bonviewaia3202689
Gruber, T. (2009). Ontology. In Springer eBooks (pp. 1963–1965). https://doi.org/10.1007/978-0-387-39940-9_1318
Munir, K., & Anjum, M. S. (2018). The use of ontologies for effective knowledge modelling and information retrieval. Applied Computing and Informatics, 14(2), 116–126. https://doi.org/10.1016/j.aci.2017.07.003
Liu, J., Chabot, Y., Troncy, R., Huynh, V., Labbé, T., & Monnin, P. (2023). From tabular data to knowledge graphs: A survey of semantic table interpretation tasks and methods. Journal of Web Semantics, 76, 100761. https://doi.org/10.1016/j.websem.2022.100761
Naqvi, M. R., Elmhadhbi, L., Sarkar, A., Archimede, B., & Karray, M. H. (2024). Survey on ontology-based explainable AI in manufacturing. Journal of Intelligent Manufacturing. https://doi.org/10.1007/s10845-023-02304-z
Osawa, H., Miyamoto, D., Hase, S., Saijo, R., Fukuchi, K., Miyake, Y. (2020). Visions of Artificial Intelligence and Robots in Science Fiction: A Computational Analysis. Int J Soc Robot. 2022;14(10):2123-2133. doi: 10.1007/s12369-022-00876-z. Epub 2022 Jul 18. PMID: 35874175; PMCID: PMC9289651
Polovina, R. (2022). On Understanding and Modelling Complex Systems, Through a Pandemic. In: Polovina, R., Polovina, S., Kemp, N. (eds) Measuring Ontologies for Value Enhancement: Aligning Computing Productivity with Human Creativity for Societal Adaptation. MOVE 2020. Communications in Computer and Information Science, vol 1694. Springer, Cham. https://doi.org/10.1007/978-3-031-22228-3_1
Polovina, R., Polovina, S. (2022). Towards Endeavor Architecture to Support Knowledge Dynamics of Societal Adaptation. In: Polovina, R., Polovina, S., Kemp, N. (eds) Measuring Ontologies for Value Enhancement: Aligning Computing Productivity with Human Creativity for Societal Adaptation. MOVE 2020. Communications in Computer and Information Science, vol 1694. Springer, Cham. https://doi.org/10.1007/978-3-031-22228-3_2
Polovina, R., Polovina, S., & Kemp, N. (Eds.). (2022). Measuring Ontologies for Value Enhancement: Aligning Computing Productivity with Human Creativity for Societal Adaptation: First International Workshop, MOVE 2020, Virtual Event, October 17–18, 2020, Revised Selected Papers (Vol. 1694). Springer Nature Switzerland. https://doi.org/10.1007/978-3-031-22228-3
Polovina, S. (2013). A Transaction-Oriented architecture for enterprise systems. International Journal of Intelligent Information Technologies, 9(4), 69–79. https://doi.org/10.4018/ijiit.2013100105
SAP. (n.d.). SAP Graph APIs. Retrieved 15 August 2024, from https://api.sap.com/graph
Schneider, T., & Šimkus, M. (2020). Ontologies and Data Management: A brief survey. KI - Künstliche Intelligenz, 34(3), 329–353. https://doi.org/10.1007/s13218-020-00686-3
Swar, A., Khoriba, G., & Belal, M. (2022). A unified ontology-based data integration approach for the internet of things. International Journal of Power Electronics and Drive Systems/International Journal of Electrical and Computer Engineering, 12(2), 2097. https://doi.org/10.11591/ijece.v12i2.pp2097-2107
Vinuesa, R., Azizpour, H., Leite, I., Balaam, M., Dignum, V., Domisch, S., Felländer, A., Langhans, S. D., Tegmark, M., & Nerini, F. F. (2020). The role of artificial intelligence in achieving the Sustainable Development Goals. Nature Communications, 11(1). https://doi.org/10.1038/s41467-019-14108-y