<p>In nanoscience, techniques based on the atomic force microscope (AFM) are a cornerstone for exploring local electrical, electrochemical, and magnetic properties at the nanoscale. As AFM's capabilities continue to evolve, the challenges of analyzing the data become more significant. With the goal of developing a prediction and clustering model for AFM electrical mode mappings based on machine learning, this work represents a step toward the analysis of big data recorded in the hyperspectral modes: AFM DataCube. To address the complexity of these multi-dimensional measurements and analysis, a self-developed tool is presented. This tool enables the analysis and processing of data, providing visualization options that include captured curves, scanned maps, animated maps in movie format, and a true 3D cube representation. In addition, the solution includes a machine learning algorithm to predict mappings from local features. In the refinement step, for prediction, the Random Forest Regressor model emerged with a root mean square error (RMSE) of 0.18, an R² value of 0.90, with an execution time of a few minutes. The clustering results are also compared to three unsupervised machine learning methods. This article presents the machine learning prediction and clustering results of DataCube, developed for all AFM DataCube modes. In this study, Scanning Spreading Resistance Microscopy (DCUBE-SSRM) is utilized to analyze a silicon-integrated microelectronic device designed for RF applications.</p>