Non-linear discriminant feature extraction using generalized back-propagation network

Abstract

SUMMARYThis paper extends linear feature extraction techniques to a wide variety of non-linear cases using a modified neural network method. This neural network method was developed by replacing the meansquare criterion with a discriminant criterion J = trace(S!,*S,). A new learning algorithm, the generalized back-propagation (GBP) algorithm, has been proposed to maximize this criterion at the network outputs. Though working in a supervised manner, the proposed learning algorithm requires no training outputs for network learning. Experimental results show that the proposed neural network method works much better than linear discriminant analysis (LDA) in linearly inseparable cases. Compared with the conventional neural network method, i.e. the standard back-propagation network (SBPN), the learning efficiency, the classification performance with limited hidden nodes and the generalization ability of the proposed network method are also more favourable. The proposed neural network method provides a promising alternative to SBPN in non-linear pattern recognition applications.

Chemometrics in China

“…Among 29 research papers published in volume 7, 1993, there were three Chinese papers, one from Professor Nianyi Chen's group of Shanghai Institute of Metallurgy, and two from Hunan University . The number of Chinese papers published in CEM gradually increased during the 1990s, accounting for five of the 49 papers published in1996. In the first decade of the millennium, this number reached a peak of nine Chinese papers among the total 68 research papers in 2009.…”

Aspects of recent developments in analytical chemometrics

“…This formulation ensures that the proposed approach provides robust results in exploratory data analysis. Second, a new neural network learning algorithm for multi-layer feed-forward network, which allows the network inputs to be updated, is proposed to accomplish NPCA in a flexible and adaptive manner.In order to enhance the feature extraction ability of clustering methods, the linear feature extraction techniques have been extended to a wide variety of non-linear cases using a modified neural network method [42] . Though working in a supervised manner, the proposed learning algorithm requires no training outputs for network learning.…”

“…In order to enhance the feature extraction ability of clustering methods, the linear feature extraction techniques have been extended to a wide variety of non-linear cases using a modified neural network method [42] . Though working in a supervised manner, the proposed learning algorithm requires no training outputs for network learning.…”

Principal Discriminant Variate Method for Classification of Multicollinear Data: Applications to Near-Infrared Spectra of Cow Blood Samples

“…After determining the rst principal discriminant variate, a 1 , based on the eigen-problem (10), one can nd successively other directions by using criterion (8) under the orthogonality constraint that a a j 5 0 (i . j).…”