提出一种求解支持向量机(SVMs)的光滑型算法.该算法基于其对偶优化模型的KKT系统,提出一类新的光滑函数族,将其KKT系统重构为一个光滑方程组,并采用光滑型算法进行求解.在适当的条件下,该算法是全局收敛和局部超线性收敛的.多个算例表明该算法非常有效,具有广阔的应用前景. A smooth algorithm for solving support vector machines (SVMs) is proposed in this paper.Based on the KKT system of its dual optimization model, a new family of smooth functions is proposed, which reconstructs its KKT system into a smooth equations, Algorithm under the appropriate conditions, the algorithm converges globally and locally superlinearly, and several examples show that the algorithm is very effective and has a broad application prospect.