Breast Cancer Diagnostic Model Based on RIME-SVM
Jiahui Ma, Siyuan Chen, Junjie Sun · 2024
Currently, breast cancer is recognized as the most prevalent illness affecting people globally. The importance of early prevention cannot be overstated, as it serves as a crucial measure in the effective treatment and management of this disease. In light of this, this paper proposes a novel breast cancer prediction model that integrates the RIME algorithm with Support Vector Machine (SVM) technology. The proposed approach involves a two-step process: initially, SVM is utilized to classify tumor cells. SVM is a powerful supervised learning model known for its efficacy in classification tasks, particularly in distinguishing between malignant and benign cells based on given input features. However, the performance of SVM heavily relies on the appropriate selection of its parameters, which can significantly impact the accuracy of the model. To address this, the RIME algorithm is employed in the second step to optimize these parameters. RIME, known for its robust optimization capabilities, iteratively adjusts the parameters of the SVM to achieve the best possible performance. Through a series of iterations, RIME fine-tunes the SVM parameters, ensuring that the model reaches its optimal configuration. The effectiveness of this combined approach is demonstrated through extensive testing. The results show that the accuracy of the test set using this RIME-optimized SVM model is approximately 3.5% higher compared to the traditional SVM training scheme. This significant improvement in accuracy underscores the potential of the proposed model in the field of breast cancer prediction, offering a more reliable method for early detection and thereby contributing to better patient outcomes.