Brain tumors rank among the most common and high-mortality cancers worldwide, typically categorized into high-grade and low-grade types. Neurosurgery remains the primary treatment method but poses significant challenges due to unclear tumor boundaries, rapid infiltration, and tumor heterogeneity. Incomplete removal of tumor tissue can lead to recurrence, while excessive removal of healthy tissue may compromise critical brain functions. Advanced technologies such as intraoperative MRI and fluorescence-guided surgery aid surgeons but have notable limitations. Hyperspectral imaging (HSI) provides a noninvasive method for distinguishing between tumor and normal tissues during surgery by utilizing spectral signatures captured in hyperspectral cubes as inputs for machine learning algorithms. This study implements and evaluates various ML techniques, including random forest, decision tree, logistic regression using metrics such as precision, recall, F1-score, accuracy, and the confusion matrix.