The problem of data dimensionality reduction is considered in the paper. The method of feature and sample selection for diagnosis model building by the precedents is proposed. It is based on feature space partitioning and evaluation of significances of terms and features. Than the measures of instance informativity are proposed to form a subsample. The example of proposed method application in the problem of biomedical data processing for chronic obstructive bronchitis diagnosis is provided.
CITATION STYLE
Subbotin, S. (2015). The instance and feature selection for neural network based diagnosis of chronic obstructive bronchitis. Studies in Computational Intelligence, 606, 215–228. https://doi.org/10.1007/978-3-319-19147-8_13
Mendeley helps you to discover research relevant for your work.