A methodology for applying machine learning to predict oil and gas reservoirs with a specified accuracy has been developed and illustrated using examples. The methodology includes principles of working with datasets, methods for interpreting the obtained models, methods for filtering local data noise, empirical methods for assessing prediction accuracy, and a method for calculating prediction accuracy based on Bayes’ theorem. This work may be of interest to specialists in the field of logic and methodology of natural sciences, petroleum and gas geology, design of geological exploration works, design of oil and gas field development, as well as specialists in data analytics and artificial intelligence in the oil and gas industry.

Vladimir A. Ostanin
Senior System Analyst
LLC “Creative”
build. 1, 84, Malygina str., Tyumen, 625026, Russia
e-mail: vost9830@mail.ru
AuthorID: 1190940

1. Zimina S.V. Vliyanie tektonicheskogo faktora na formirovanie treshchinno-porovogo tipa kollektora na primere gorizonta YU1 Dvurechenskogo mestorozhdeniya nefti Tomskoi oblasti [The influence of the tectonic factor on the formation of a fractured-pore type of reservoir using the example of the Yu1 horizon of the Dvurechenskoye oil field in the Tomsk region]: dissertation for the degree of candidate of geological and mineralogical sciences. Tyumen’: TyumGNGU; 2010.167 p.
2. Ostanin V.A. Perspektivy ispol’zovaniya avtomaticheskogo mashinnogo obucheniya v zadachakh kompleksnoi interpretatsii geofizicheskikh dannykh na primere raionirovaniya Tomskoi oblasti po veroyatnoi neftegazonosnosti [Prospects for using automatic machine learning in problems of complex interpretation of geophysical data using the example of zoning the Tomsk region according to probable oil and gas content]. In: Perspektivy razvitiya nauki v sovremennom mire. Sbornik nauchnykh statei po materialam XII Mezhdunarodnoi nauchno-prakticheskoi konferentsii (Ufa, 7 April 2023.). Pt. 2. Ufa: Vestnik nauki; 2023. pp. 137–156.
3. Sedukhin O. Interpretatsiya modelei i diagnostika sdviga dannykh: LIME, SHAP i Shapley Flow [Model interpretation and data drift diagnostics: LIME, SHAP and Shapley Flow]. In: Open Data Science. 13.01.2022. Available at: https://habr.com/ru/companies/ods/articles/599573/ (accessed 25.01.2024).
4. Chernobrovov A.I. Interpretirui ehto: metod SHAP v Data Science [Interpret it: SHAP method in Data Science]. 21.06.2020. Available at: https://chernobrovov.ru/articles/interpretiruj-eto-metod-shap-v-data-science.html (accessed 25.01.2024).