This study presents our experiments with ChatGPT aimed at constructing a prototype domain ontology for World Heritage. We relied on the official United Nations Educational, Scientific and Cultural Organization classification published within the World Heritage Convention and its 2023 operational guidelines document as our trusted data sources. Our focus encompassed four main tasks: (i) gathering competency questions; (ii) constructing classes along with their taxonomy and definitions; (iii) populating the ontology with automatically generated instances; and (iv) evaluating the ontology through SPARQL queries generated from competency questions. By employing a naive approach paired with in-context learning and some prompt engineering techniques, our goal is to assess the capabilities and limitations of ChatGPT in supporting the construction of prototype ontologies by nonexperts or those with limited domain knowledge for application demonstration proposals. The results indicate that while this exciting area can expedite the ontology learning process, it also encounters certain technical limitations in uncovering implicit and less explicit semantics in input data. Additionally, proficient prompting skills are required to guide the model for producing accurate results.
AlKhamissiB.LiM.CelikyilmazA.DiabM.GhazvininejadM. (2022). A review on language models as knowledge bases. arXiv preprint, arXiv:220406031. https://doi.org/10.48550/arXiv.2204.06031
2.
AmirM.BaruahM.EslamialishahM.EhsaniS.BahramaliA.Naddaf-ShS., et al. (2023). Truveta Mapper: A zero-shot ontology alignment framework. arXiv preprint, arXiv:230109767. https://doi.org/10.48550/arXiv.2301.09767
3.
ArpR.SmithB.SpearA. D. (2015). Building ontologies with basic formal ontology. MIT Press.
4.
AvilaC. V. S.VidalV. M.FrancoW.CasanovaM. A. (2024). Experiments with text-to-SPARQL based on ChatGPT. In 2024 IEEE 18th international conference on semantic computing (ICSC) (pp. 277–284). IEEE.
5.
BaaderF.HorrocksI.SattlerU. (2005). Description logics as ontology languages for the semantic web. In Mechanizing mathematical reasoning: Essays in honor of Jörg H. Siekmann on the occasion of his 60th birthday (pp. 228–248). Springer. https://doi.org/10.1007/978-3-540-32254-2_14
6.
Babaei GiglouH.D’SouzaJ.AuerS. (2023). LLMs4OL: Large language models for ontology learning. In: International semantic web conference (pp. 408–427). Springer.
7.
BachS. H.SanhV.YongZ. X.WebsonA.RaffelC.NayakN. V., et al. (2022). Promptsource: An integrated development environment and repository for natural language prompts. arXiv preprint, arXiv:220201279. https://doi.org/10.48550/arXiv.2202.01279
8.
BiesterF.GaudioD. D.AbdelaalM. (2023). Enhancing knowledge base construction from pre-trained language models using prompt ensembles. In KBC-LM/LM-KBC@ISWChttps://api.semanticscholar.org/CorpusID:266364474
9.
BizerC.HeathT.Berners-LeeT. (2008). Linked data: Principles and state of the art. In: World wide web conference (Vol. 1, p. 40). https://www.w3.org/2008/Talks/WWW2008-W3CTrack-LOD.pdf
10.
BlaneyJ. (2017). Introduction to the principles of linked open data. The Programming Historian, 6. https://doi.org/10.46430/phen0068
11.
BodenreiderO. (2004). The unified medical language system (UMLS): Integrating biomedical terminology. Nucleic Acids Research, 32(suppl_1), D267–D270. https://doi.org/10.1093/nar/gkh061
12.
CarrieroV. A.GangemiA.MancinelliM. L.MarinucciL.NuzzoleseA. G.PresuttiV., et al. (2019). ArCo: The Italian cultural heritage knowledge graph. In: The semantic web–ISWC 2019: 18th international semantic web conference, Auckland, New Zealand, October 26–30, 2019, Proceedings, Part II 18 (pp. 36–52). Springer.
13.
CartaS.GiulianiA.PianoL.PoddaA. S.PompianuL.TiddiaS. G. (2023). Iterative zero-shot llm prompting for knowledge graph construction. arXiv preprint, arXiv:230701128. https://doi.org/10.48550/arXiv.2307.01128
14.
CaufieldJ. H.HegdeH.EmonetV.HarrisN. L.JoachimiakM. P.MatentzogluN., et al. (2024). Structured prompt interrogation and recursive extraction of semantics (SPIRES): A method for populating knowledge bases using zero-shot learning. Bioinformatics (Oxford, England), 40(3), btae104. https://doi.org/10.1093/bioinformatics/btae104
15.
ChenJ.HeY.GengY.Jiménez-RuizE.DongH.HorrocksI. (2023). Contextual semantic embeddings for ontology subsumption prediction. World Wide Web, 26(5), 2569–2591. https://doi.org/10.1007/s11280-023-01169-9
16.
ClaviéBCiceuANaylorFSouliéGBrightwellT. (2023). Large language models in the workplace: A case study on prompt engineering for job type classification. In: International conference on applications of natural language to information systems (pp. 3–17). Springer.
17.
Committee for the Protection of the World Cultural I, Natural Heritage. (2024). Operational guidelines for the implementation of the world heritage convention. UNESCO: United Nations Educational, Scientific and Cultural Organisation. https://whc.unesco.org/en/guidelines/
DevlinJ.ChangM.LeeK.ToutanovaK. (2019). Pre-training of deep bidirectional transformers for language understanding. In Proceedings of the 2019 conference of the North American chapter of the association for computational linguistics: Human language technologies, vol. 1 (long and short papers) (pp. 4171–4186). Association for Computational Linguistics.
21.
DingN.HuS.ZhaoW.ChenY.LiuZ.ZhengH. T.SunM. (2021). Openprompt: An open-source framework for prompt-learning. arXiv preprint, arXiv:211101998. https://doi.org/10.48550/arXiv.2111.01998
22.
EhrigM. (2006). Ontology alignment: Bridging the semantic gap (Vol. 4). Springer Science & Business Media.
23.
El IdrissiB.BaïnaS.BaïnaK. (2013). Automatic generation of ontology from data models: A practical evaluation of existing approaches. In: IEEE 7th international conference on research challenges in information science (RCIS) (pp. 1–12). IEEE.
24.
Fernández-LópezM.Gómez-PérezA.Juristo JuzgadoN. (1997). Methontology: From ontological art towards ontological engineering. In: Proceedings of the ontological engineering AAAI-97 Spring symposium series (pp. 33-40). American Association for Artificial Intelligence; Ontology Engineering Group – OEG. https://oa.upm.es/5484/
25.
FreyJ.MeyerL. P.ArndtN.BreiF.BulertK. (2023). Benchmarking the abilities of large language models for RDF knowledge graph creation and comprehension: How well do LLMS speak turtle? arXiv preprint, arXiv:230917122. https://doi.org/10.48550/arXiv.2309.17122
Gómez-PérezA.Manzano-MachoD., et al. (2003). A survey of ontology learning methods and techniques. OntoWeb Deliverable D, 1(5), 1–86.
28.
GordonA. D. (1999). Classification. CRC Press.
29.
GuY.TinnR.ChengH.LucasM.UsuyamaN.LiuX., et al. (2021). Domain-specific language model pretraining for biomedical natural language processing. ACM Transactions on Computing for Healthcare (HEALTH), 3(1), 1–23. https://doi.org/10.1145/3458754
30.
GuhaR. V.BrickleyD.MacbethS. (2016). Schema. org: Evolution of structured data on the web. Communications of the ACM, 59(2), 44–51. https://doi.org/10.1145/2844544
31.
HeY.ChenJ.AntonyrajahD.HorrocksI. (2022). BERTMap: A BERT-based ontology alignment system. In Proceedings of the AAAI conference on artificial intelligence (Vol. 36, pp. 5684–5691). AAAI Press.
32.
HeY.ChenJ.DongH.HorrocksI. (2023). Exploring large language models for ontology alignment. arXiv preprint, arXiv:230907172. https://doi.org/10.48550/arXiv.2309.07172
33.
HeY.ChenJ.Jiménez-RuizE.DongH.HorrocksI. (2023). Language model analysis for ontology subsumption inference. arXiv preprint, arXiv:230206761. https://doi.org/10.48550/arXiv.2302.06761
34.
HertlingS.PaulheimH. (2023). OLaLa: Ontology matching with large language models. In Proceedings of the 12th knowledge capture conference 2023 (pp. 131–139). ACM.
35.
HuaS. (2010). World heritage classification and related issues—A case study of the “convention concerning the protection of the World cultural and natural heritage”. Procedia-Social and Behavioral Sciences, 2(5), 6954–6961. https://doi.org/10.1016/j.sbspro.2010.05.048
36.
HyvönenE. (2020). Linked open data infrastructure for digital humanities in Finland. In: Digital humanities in the Nordic countries (pp. 254–259). CEUR. http://ceur-ws.org/Vol-2612/
37.
HyvönenE.ViljanenK.TuominenJ.SeppäläK. (2008). Building a national semantic web ontology and ontology service infrastructure—the FinnONTO approach. In: The semantic web: Research, applications: 5th European semantic web conference, ESWC 2008, Tenerife, Canary Islands, Spain, June 1–5, 2008 Proceedings 5 (pp. 95–109). Springer.
38.
KommineniV. K.König-RiesB.SamuelS. (2024). From human experts to machines: An LLM supported approach to ontology and knowledge graph construction. arXiv preprint, arXiv:240308345. https://doi.org/10.48550/arXiv.2403.08345
39.
KotisK. I.VourosG. A.SpiliotopoulosD. (2020). Ontology engineering methodologies for the evolution of living and reused ontologies: Status, trends, findings and recommendations. The Knowledge Engineering Review, 35, e4. https://doi.org/10.1017/S0269888920000065
40.
KumarG. T.SinghA. K. (2022). Intangible cultural heritage: Knowledge classification and management. Informatics Studies, 9(2), 9–24.
41.
LiT.HuangW.PapasarantopoulosN.VougiouklisP.PanJ. Z. (2022). Task-specific pre-training and prompt decomposition for knowledge graph population with language models. arXiv preprint, arXiv:220812539. https://doi.org/10.48550/arXiv.2208.12539
42.
LissaM.BhuvaneswariV.DeviT.KumarJ. S.RajeswariR. (2022). Semantic data model for knowledge representation and dissemination of cultural heritage site, Poompuhar. Current Science, 123(10), 1237. https://doi.org/10.18520/cs/v123/i10/1237-1245
43.
LiuH.PerlY.GellerJ. (2020). Concept placement using BERT trained by transforming and summarizing biomedical ontology structure. Journal of Biomedical Informatics, 112, 103607. https://doi.org/10.1016/j.jbi.2020.103607
44.
MareeM.RattroutA.AltawilM.BelkhatirM. (2021). Multi-modality search and recommendation on Palestinian cultural heritage based on the holy-land ontology and extrinsic semantic resources. Journal on Computing and Cultural Heritage (JOCCH), 14(3), 1–23. https://doi.org/10.1145/3447523
45.
MateiuP.GrozaA. (2023). Ontology engineering with large language models. arXiv preprint, arXiv:230716699. https://doi.org/10.48550/arXiv.2307.16699
46.
MeyerL. P.StadlerC.FreyJ.RadtkeN.JunghannsK.MeissnerR., et al. (2023). LLM-assisted knowledge graph engineering: Experiments with ChatGPT. In Working conference on artificial intelligence development for a resilient and sustainable tomorrow (pp. 103–115). Springer.
ProckoT. T.OchoaO.ElviraT. (2023). Automatic generation of BFO-compliant Aristotelian definitions in OWL ontologies with GPT. In: 2023 fifth international conference on transdisciplinary AI (TransAI) (pp. 141–146). IEEE.
51.
QiangZ.WangW.TaylorK. (2023). Agent-OM: Leveraging large language models for ontology matching. arXiv preprint, arXiv:231200326. https://doi.org/10.48550/arXiv.2312.00326
52.
SattarA.SurinE. S. MAhmadM. N.AhmadM.MahmoodA. K. (2020). Comparative analysis of methodologies for domain ontology development: A systematic review. International Journal of Advanced Computer Science and Applications, 11(5).10.14569/IJACSA.2020.0110515
53.
StrakováJ.FučíkováE.HajičJ.UrešováZ. (2023). Extending an event-type ontology: Adding verbs and classes using fine-tuned LLMs suggestions. In J. Prange & A. Friedrich (Eds.), Proceedings of the 17th linguistic annotation workshop (LAW-XVII) (pp. 85–95). Association for Computational Linguistics. https://aclanthology.org/2023.law-1.9
WiśniewskiD.PotoniecJ.ŁawrynowiczA.KeetC. M. (2019). Analysis of ontology competency questions and their formalizations in SPARQL-OWL. Journal of Web Semantics, 59, 100534. https://doi.org/10.1016/j.websem.2019.100534
60.
YangS.HouM. (2023). Knowledge graph representation method for semantic 3D modeling of Chinese grottoes. Heritage Science, 11(1), 266. https://doi.org/10.1186/s40494-023-01084-2
61.
ZhangB.CarrieroV. A.SchreiberhuberK.TsanevaS.GonzálezL. S.KimJ., et al. (2024). OntoChat: A framework for conversational ontology engineering using language models. arXiv preprint, arXiv:240305921. https://doi.org/10.48550/arXiv.2403.05921
62.
ZhengY.SeppänenO.SeissS.MelznerJ. (2023). Testing ChatGPT-aided SPARQL generation for semantic construction information retrieval. In International conference on construction applications of virtual reality (pp. 751–760). Firenze University Press.
63.
ZhouY.MuresanuA. I.HanZ.PasterK.PitisS.ChanH., et al. (2022). Large language models are human-level prompt engineers. arXiv preprint, arXiv:221101910. https://doi.org/10.48550/arXiv.2211.01910
