Cancer of the breast is one of the deadliest diseases encountered by women and requires early diagnosis. Although more time is required for the traditional diagnostic process, a realistic and straightforward approach may be achieved through machine learning approaches to detect ailment. Advances in technology, however, generate different kinds of high-dimensioned data, majorly cancer or medical data. Staggeringly high data makes it harder for them to obtain insight knowledge; unrepresentative knowledge can contribute to skewed outcomes in classification. The feature selection process may be used to enhance classification performance to solve any of these challenges. In this article, Particle Swarm Optimization (PSO) is suggested to optimize the efficiency of the classification with the Decision Tree algorithm on a Wisconsin Breast Cancer dataset. The findings reveal that the performance of the system was 92.26% accuracy likened to the state-of-the-art. In conclusion, the system will help minimize the existence of breast cancer disease by creating an early detection system for diagnosis based on a machine learning approach, and it will help health practitioners in decision-making.