Classifier

The problem of convergence of the procedure for synthesizing classifier schemes in the methods of logical and algorithmic classification trees is considered. An upper estimate of the complexity of the algorithm tree scheme is proposed in the problem of approximating an array of real data with a set of generalized features with a fixed criterion for stopping the branching procedure at the stage of constructing a classification tree.

In the paper, the concept of the formation of the national spatial data infrastructure (NSDI) in Ukraine has been investigated, the complex NSDI industry standards and the classes of objects of the urban-planning cadaster have been analyzed.

The article describes the research of the peculiarities of methods and algorithms for the recognition of mathematical expressions. The possibility of simultaneous execution of structural analysis and classification of characters is investigated. The process of classification of the symbols and construction of the corresponding system, based on methods of machine learning, is described. The developed iterative algorithm is implemented in the design of the intelligent information system for the recognition of mathematical expressions.

In the paper, a classification method is proposed. It is based on Combined Swarm Negative Selection Algorithm, which was originally designed for binary classification problems. The accuracy of developed algorithm was tested in an experimental way with the use of microarray data sets. The experiments confirmed that direction of changes introduced in developed algorithm improves its accuracy in comparison to other classification algorithms.