Generalized re-weighting local sampling mean discriminant analysis

摘要

Despite the general success in the pattern recognition community, linear discriminant analysis (LDA) has four intrinsic drawbacks. In this paper, we propose a new feature extraction algorithm, namely, local sampling mean discriminant analysis (LSMDA), to make up for the first three drawbacks, and a generalized re-weighting (GRW) framework to make up for the fourth drawback. Extensive experiments are conducted on both synthetic and real-world datasets to evaluate the classification performance of our work. The experimental results demonstrate the effectiveness of both LSMDA and the GRW framework in classifications.