Comparison of Different Knowledge Representations for Complex Structured Objects in Solving AI Problems
Abstract
The problem of knowledge representation for a complex structured object is one of the actual problems of AI. This is due to the fact that many of the objects under study are not a single indivisible object characterized by its properties, but complex structures whose elements have some known properties and are in some, often multiplace, relations with each other.
An approach to the representation of such knowledge based on first-order logic (predicate calculus formulas) is compared in this paper with two currently widespread approaches based on the representation of data information with the use of finite-valued strings or graphs.
It is shown that the use of predicate calculus formulas for description of a complex structured object, despite the NP-difficulty of the solved problems arising after formalization, actually have no greater computational complexity than the other two approaches, what is usually not mentioned by their supporters.
An algorithm for constructing an ontology is proposed that does not depend on the methodof desc ribing an object, and is based on the selection of the maximum common property of objects from a given set.
References
G. Bellinger, D. Casstro, and A. Mills, “Date, Information, Knowledge, and Wisdom,” in Systems-thinking.org, 2004. [Online]. Available: https://systems-thinking.org/dikw/dikw.htm
A. J. Godbout and G. M. Godbout, “Filtering Knowledge: Changing Information into Knowledge Assets,” Journal of Systemic Knowledge Management, vol. 1, no. 3, 1999.
H. J. van Meter, “Revising the DIKW Pyramid and the Real Relationship Between Data, Information, Knowledge and Wisdom,” Law, Technology and Humans, vol. 2, no. 2, pp. 69–80, 2020; doi: 10.5204/lthj.1470
P. Cooper, “Data, information, knowledge and wisdom,” INFORMATICS, vol. 18, no. 1, pp. 55–56, 2017; doi: https://doi.org/10.1016/j.mpaic.2016.10.006
A. Darwiche, “Three Modern Roles for Logic in AI,” in Arxiv.org, 2020. [Online]. Available: https://arxiv.org/pdf/2004.08599.pdf
A. Amarilli, “Provenance in Databases and Links to Knowledge Compilation,” in KOCOON workshop on knowledge compilation, Arras, France, Dec, 17, 2019, 2019. [Slides]. Available: https://a3nm.net/work/talks/kocoon2019/amarilli2019provenance_slides.pdf
M. Gelfond, “Logic Based Approach to AI,” in fmfi-uk.hq.sk, 2005. [Online]. Available: https://fmfi-uk.hq.sk/Informatika/Uvod%20Do%20Umelej%20Inteligencie/clanky/logic-ai.pdf
R. Calegari, G. Ciatto, E. Denti, and A. Omicini, “Logic-Based Technologies for Intelligent Systems: State of the Art and Perspectives,” Information, vol. 11, no. 3, p. 167, 2020; doi: 10.3390/info11030167
V. S. Gurin, E. V. Kostrov, Yu. Yu. Gavrilenko, D. F. Saada, E. A. Il’yushin, and I. V. Chizhov, “Knowledge Graph Essentials and Key Technologies,” Modern Information Technologies and IT-Education, vol. 15, no. 4, pp. 932–944, 2019 (in Russian); doi: 10.25559/SITITO.15.201904.932-944
Z. Zhang, P. Cui, and W. Zhu, “Deep Learning on Graphs: A Survey,” Journal of latex class files, vol. 14, no. 8, pp. 1–24, 2015.
S. V. Mikoni, “On the Quality of Ontological Models,” Ontology of Designing, vol. 7, no. 3, pp. 347–360, 2017; doi: 10.18287/2223-9537-2017-7-3-347-360
N. F. Noy and D. L. McGuinness, “Ontology Development 101: A Guide to Creating Your First Ontology,” Technical Report KSL-01-05 and Stanford Medical Informatics Technical Report SMI-2001-0880, Stanford, CA, USA: Stanford Knowledge Systems Laboratory, 2001.
A. Mykhailiuk and M. Petrenko, “Machine learning and ontologies as two approaches for building intellectual information systems,” International Journal "Information Technologies & Knowledge", vol. 13, no. 1, pp. 55–75, 2019 (in Russian).
N. G. Nilsson Problem-Solving Methods in Artificial Intelligence, Moscow: Mir, 1973 (in Russian).
T. M. Kosovskaya, “Dokazatel’stva otsenok chisla shagov resheniya nekotorykh zadachraspoznavaniya obrazov, imeyushchikh logicheskie opisanii” [Proofs of the number of steps bounds for solving of some pattern recognition problems with logical description], Vestnik of St-Petersburg University. Ser. 1, no. 4, pp. 82–90, 2007 (in Russian).
T. M. Kosovskaya, “Fuzzy Recognition by Logic-Predicate Network,” Adv. Sci. Technol. Eng. Syst. J., vol. 5, no. 4, pp. 686–699, 2020; doi: 10.25046/aj050482
T. M. Kosovskaya and D. A. Petrov, “Extraction of a maximal common sub-formula of predicate formulas for the solving of some artificial intelligence problems,” Vestnik S.-Petersburg Univ. Ser. 10. Prikl. Mat. Inform. Prots. Upr., vol. 13, no. 3, pp. 250–263, 2017 (in Russian); doi: 10.21638/11701/spbu10.2017.303
T. M. Kosovskaya, “Isomorphism of predicate formulas in artificil intelligence problems,” Zap. Nauchn. Sem. POMI, vol. 499, pp. 38–52, 2021 (in Russian).
C. M. Bishop, Pattern Recognition and Machine Learning, New York: Springer, 2006.
J. Oh, K. Yun, U. Maoz, T.-S. Kim, and J.-H. Chaea, “Identifying depression in the National Health and Nutrition Examination Survey data using a deep learning algorithm,” Journal of Affective Disorders, vol. 257, pp. 623–631, 2019; doi: 10.1016/j.jad.2019.06.034
S. J. Russel and P. Norvig, Artificial Intelligence. A Modern Approach, Moscow: Izdatelskii dom “Vil’yams”, 2006 (in Russian).
M. R. Garey and D. S. Johnson, Computers and Intractability: A Guide to the Theory of NP-Completeness, Moscow: Mir, 1982 (in Russian).
T. M. Kosovskaya, “Isomorphous Subformules Isolation as a Means to Construct Logical Ontology,” in Proc. 13th Russian Multiconference on Management Problems. "Information technology management (ITU-2020), 7–8 Oct. 2020, St. Petersburg, St. Petersburg, Russia: Concern CSRI Elektropribor, 2020, pp. 84–86 (in Russian).
This work is licensed under a Creative Commons Attribution 4.0 International License.
