Abstract
The problem of estimating a classification rule with partially classified observations, which often occurs in biological and ecological modelling, and which is of major interest in pattern recognition, is discussed. Radial basis function networks for classification problems are presented and compared with the discriminant analysis with partially classified data, in situations where some observations in the training set are unclassified. An application on a set of morphometric data obtained from the skulls of 288 specimens of Microtus subterraneus and Microtus multiplex is performed. This example illustrates how the use of both classified and unclassified observations in the estimate of the hidden layer parameters has the potential to greatly improve the network performances.
Highlights
One of the major problems related to practical applications in pattern recognition is the presence of partially classified data
This work is an attempt to explain and illustrate the use of Radial basis function (RBF) networks in situations where partially classified data sets occur and to show the differences between this methodology and other competitive methods which are often used in these situations
The goal of this paper is to make RBF networks more popular, since they appear to be rather less well known than the classical multi-layer perceptron, in the neural networks field, and than discriminant analysis and discriminant analysis with partially classified observations, in statistics
Summary
One of the major problems related to practical applications in pattern recognition is the presence of partially classified data. In these situations the population from which the sample is taken consists itself of a number of several homogeneous sub-populations, but the group membership of the training data is known only for some input vectors. If the quantity of data available is sufficiently large, and the proportion of unclassified observations is small, the simplest solution is to discard those patterns from the data set This approach, is implicitly assuming that the cause of the omission of the group membership is independent of the data itself. The network performances are measured in terms of classification error rate and generalisation to unobserved patterns
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
