Evaluation Model of the Artist Based on Fuzzy Membership to Improve the Principal Component Analysis of Robust Kernel

Principal component analysis is a number of related indexes into a multivariate statistical method a fewrelated indicators, often used in data compression and feature extraction are widely used in industry, agriculture, economy, biology, medicine, astronomy, geography and other fields. In the classical principal component analysis, each training data in construction the main ingredient is the same. However, in many practical problems, the significance and effect of the training data is different, usually some of the data than other data is moreimportant. We should pay more attention to the important data, should play a greater role in the construction of the main components, and the data may be not credible is the abnormal data, should limit its role. In this paper, each training data gives a confidence weight to the training data as fuzzy points in the sample space, based on the research Principal component analysis and kernel principal component analysis of fuzzy point data. In this paper, an analysis method based on the principal component analysis of the objective weight is presented, and the method is applied to the evaluation of the value of the artist's creation. The paper analyses the in kernel principal component analysis KPCA, input data space X reflect the shoot diameter: R P, h is projected onto a new high-dimensional feature space h after, although you can achieve nonlinear feature extraction, but still in the presence of outliers. In feature space, we can reduce the effect of outliers, and avoid the effect of the traditionalprinciple component analysis, so it has the advantages of robust and nonlinear. In fact, we show that the fuzzy membership degree of fuzzy membership degree is improved, and the accuracy rate is 84%.We use the kernel principal component analysis method to improve the original foundation, and realize the construction of the artist evaluation model, and improve the accuracy.