As a kind of supervised learning algorithm, ridge regression algorithm has a wide range of applications. A quantum ridge regression algorithm is proposed by combining quantum singular value estimation with classical ridge regression algorithm. In the proposed algorithm, the parallel property of quantum computation is utilized to solve the fitting parameters of ridge regression and obtain the predicted values. Complexity analysis shows that the proposed algorithm effectively solves the problem of matrix expansion or matrix operation when the data matrix is non-Hermitian matrix, and has exponential acceleration in running time compared with the classical algorithms. In addition, the quantum circuit diagram of the proposed algorithm is also provided and the key steps of the algorithm are simulated. The simulation results confirm its effectiveness and feasibility.