Jamshidi, AminAkbay, Deniz2026-02-032026-02-0320252228-7817https://doi.org/10.22059/geope.2025.390443.648807https://hdl.handle.net/20.500.12428/34267The most important criteria needed for the investigation and characterization of a rock mass on site in a geotechnical project are its uniaxial compressive strength (UCS) and tensile strength (TS). The UCS and TS of rocks are determined directly by complex laboratory or field tests that require specialized prepared samples and equipment. Therefore, the UCS and TS of rocks are estimated through several index parameters via regression analysis. The point load index (PLI) due to its simplicity and quickness is a common parameter for estimating the UCS and TS of rocks. In this study, data mining tools are used to estimate the UCS and TS [determined through the Brazilian tensile strength (BTS) test] of rock using PLI. The statistical parameters, including mean absolute error (MAE), root mean squared error (RMSE), and correlation coefficient (r), are used to evaluate the performance of each data mining tool. The validity and accuracy of platforms' data mining tools were verified according to the statistical parameters. The results indicated that all three platforms' data mining tools exhibited remarkable ability to predict UCS and BTS using PLI. Finally, using platforms' data mining tools obviates the need to perform the UCS and BTS tests as time-consuming and laborious efforts. This article is an open-access article distributed under the terms and conditions of the Creative Commons Attribution (CC-BY) license.eninfo:eu-repo/semantics/closedAccessBrazilian Tensile StrengthMachine Learning.Regression AnalysisUniaxial Compressive StrengthArticle15240952710.22059/geope.2025.390443.6488072-s2.0-105025566406Q3