Breast cancer is the most common cancer among women worldwide as well as one of the leading causes of cancer death. Therefore, timely and accurate prediction of the disease is essential to improve survival. Early detection and intervention can significantly improve patient survival rates, improve quality of life, and reduce overall medical costs. This study employs machine-learning algorithms to analyze the Breast Cancer Wisconsin Data Set, predicting whether benign or malignant cases. These algorithms include Support Vector Machine (SVM), Random Forest, and Neural Network. The research found that the Neural Network model was particularly effective in identifying malignant cases minimizing missed diagnoses, while the Random Forest model excelled in reducing false positives. The SVM provided a balanced approach, minimizing both false positives and negatives. The purpose of this paper is to prove the potential of machine learning to enhance early breast cancer diagnosis, improving treatment outcomes and survival rates.