The shape and location of cutouts in sandwich structures are important factors that can greatly affect the natural frequencies of these structures. Cutouts can change the overall stiffness and mass distribution of the structure, which can result in changes to the natural frequencies. Cutout shape can also affect how the structure deforms under load, which can further impact the natural frequencies. The location of the cutout is also critical, as it can change the way the load is distributed throughout the structure. So, understanding how the shape and location of cutouts affect the natural frequencies of sandwich structures is important for the design and optimization of these structures. It can help engineers to minimize the influence of cutouts on the natural frequencies and ensure that the structures perform as desired. Using machine learning techniques to classify cutout shapes and predict cutout locations can also help engineers to better understand these effects, and allow them to make more informed decisions during the design process. Therefore, this study investigates the classification of cutout shapes and the prediction of the cutout locations using different machine-learning methods. Natural frequencies of sandwich structures are obtained using the finite element method to use as input to machine learning methods. Cutout shapes and cutout locations are used as output for classification and regression studies respectively. From the result, it was obtained that the cutout shape was classified with an accuracy of 99.5 %. Also, the cutout location is predicted with an RMSE value of 0.000090883.