The response of municipal solid waste (MSW) for static and dynamic loading varies from region to region and must be studied uniquely across sites. Tradition laboratory investigation on MSW poses challenges related to unpleasant odors due to degradation, which makes it difficult and time-consuming. To tackle these problems, predictive models can be leveraged to estimate the shear modulus (G) of the MSW. This work used boosting-based ensemble machine learning (ML) methods in combination with explainable artificial intelligence (XAI) techniques to study the dynamic behavior of MSW. A vast database of 153 cyclic triaxial tests was collected from existing literature. The age of MSW, shear strain (ShS), confining pressure (CP), unit weight of MSW (UW), plastic content (POP), and loading frequency (F) were considered as input parameters. The prediction of shear modulus was performed using six boosting-based ensemble methods. The model explainability techniques are utilized to study the influence of input features on the shear modulus of MSW. The results indicate that the ensemble method extreme gradient boosting (XGBoost) emerges as the best-performing model in both the training and testing phases. Furthermore, the impact of key variables on the prediction of shear modulus was performed using sensitivity analysis.