The optimization of least-squares filter design can be formulated as an eigenproblem by solving an appropriate positive-definite matrix. Based on Rayleigh's principle, eigenfilter design can be achieved by solving a single eigenvector corresponding to the smallest eigenvalue of an associated matrix. In this paper, the minor component analysis based on neural approach is exploited for the design of eigenfilter with effectiveness. As the learning algorithm achieves convergence, the weight vector of the neural system would approximate to the minimum eigenvector which results in the optimal filter coefficients of the eigenfilter design. Simulation results indicate that the proposed neural learning approach achieves good performance.