<p>This study investigated the prediction of the unconfined compressive strength of rocks using several rock properties, including point load strength index, Schmidt rebound hardness, block punch index, porosity, and specific gravity. Numerous machine learning models were applied, including linear regression, polynomial regression, decision tree regression, support vector regression, random forest regression, artificial neural networks, convolutional neural networks, and long short-term memory networks. The long short-term memory model showed the best predictive performance, achieving a coefficient of determination of 0.92 and a low mean absolute error of 10.94 on the test dataset. Sensitivity analysis indicated that the point load strength index was the most influential predictor across all models. Correlation analysis showed strong positive relationships between unconfined compressive strength and factors such as Schmidt rebound hardness, block punch index, and specific gravity, with correlation coefficients of 0.928, 0.88, and 0.884, respectively. These connections underscore the significance of these variables in affecting rock strength. The long short-term memory model performed exceptionally well in predicting unconfined compressive strength, with great accuracy, consistency, and adaptability. In geotechnical engineering, this approach provides a reliable and efficient substitute for traditional methods of assessing rock strength.</p>