In 2022, cancer was responsible for nearly 10 million deaths worldwide. Traditional treatments such as surgery, radiotherapy, and chemotherapy have been the primary approaches; however, they are not without limitations. Matrixmetalloproteinase 9 (MMP-9) has been an attractive target for inhibition in anticancer therapy. Machine learning (ML) and its derivatives hold potential to advance drug discovery. In this paper, we will employ the moth-flameoptimization (MFO) algorithm, coupled with support vector machine (SVM), to predict MMP-9 inhibitors as anticancer agents and evaluate SVM’s performance in this context. MFO will be used to select relevant feature to reduce computational costs, while SVM is noted for its capability to handle high-dimensional data. This research aims to achieve improved predictive accuracy, thus promoting more efficient computational methods in drug discovery.