Hepatocellular Carcinoma (HCC), the most common form of liver cancer, has a rising mortality rate primarily due to a lack of awareness about prevention strategies. This work addresses these challenges by utilizing machine learning techniques such as feature selection, classification, and preprocessing to improve early diagnosis. A novel approach is proposed, which combines a new feature selection method with advanced preprocessing to detect diseases in their early stages. Specifically, the K-Nearest Neighbor with May Fly (KNN-MF) technique is used for preprocessing to enhance accuracy. For feature selection, a hybrid Remora with Whale Optimization Algorithm (RWOA) is introduced to minimize time complexity. Classification is performed using Support Vector Machines (SVM). Experimental results demonstrate that the MF-RWOA technique consistently achieves an accuracy rate of 97.2%. Key evaluation metrics include accuracy, precision, recall, f-measure, and processing time. In future work, we plan to explore ensemble classification methods across various healthcare datasets to further improve performance and outcomes.