Artificial intelligence

Abstract: The work is dedicated to describing a methodology for generating activity scenarios based on causal networks formed using generative artificial intelligence. The methodology is based on the use of a bidirectional algorithm for generating causal networks. According to this algorithm, two networks are formed and then combined – the first network starts from a node corresponding to the initial state of the problem (the root cause), and the second network corresponds to the goal that needs to be achieved. The article demonstrates the possibility of constructing such causal networks based on the use of a generative transformer like ChatGPT, and provides an example of scenario generation in the subject field of mobile communication. The methodology combines tools for text analysis and the formation of causal networks, followed by the selection and ranking of narrative chains based on the analysis of these networks.

Keywords: ChatGPT, causal networks, domain models, generative artificial intelligence, scenario analysis

References:

1. St. Wolfram. What Is ChatGPT Doing ... and Why Does it Work? – Wolfram Media, Inc. March 9, 2023. 112 p. ISBN-10: ‎ 1579550819
2. Bellman, Richard (1957), Dynamic Programming, Princeton University Press. Dover paperback edition (2003), ISBN: 0-486-42809-5
3. Ken Cherven. “Mastering Gephi Network Visualization”. – Packt Publishing, 2015. – 378 p.
4. Lande, Dmytro and Strashnoy, Leonard. GPT Semantic Networking: A Dream of the Semantic Web – The Time is Now. – Kyiv: Engineering, 2023. 168 p. ISBN: 978-966-2344-94-3. URL: (https://bigsearch.space/ datasets/Lande_Str_Book.pdf)
5. Lande, Dmitry; Strashnoy, Leonard.Hierarchical Formation of Causal Networks Based on ChatGPT. Available at SSRN: https://papers.ssrn.com/sol3/papers.cfm?abstract_id=4440629, DOI: https://dx.doi.org/10.2139/ssrn.4440629 (May 8, 2023). - 13 p.